نشریه پژوهش های اقلیم شناسی
پیاپی 50 (تابستان 1401)

One of the consequences of the increase and accumulation of greenhouse gases in the atmosphere is kown as global warming, which is undoubtedly one of the most important environmental challenges in the world, especially in the Middle East. Given the scarcity of water resources in recent years, the consequences of global warming and climate change in various countries have reached a very worrying level. Carbon dioxide and Methane are known as two of the most important human greenhouse gases in the atmosphere, accounting for 64% and 18%, respectively, of long-lived radiation induction (LLGHGs) (Forster et al, 2007). Methane is considered as the second most important anthropogenic greenhouse gas after Carbon Dioxide.The most important sources of Methane emissions include: biomass incineration, artificial human emissions, wetland emissions, and wastes. Despite the importance of Methane for physical and atmospheric conditions, the spatial distribution of global resources and Methane sinks is not well understood. With the launch of Methane measurement from satellites, knowledge about the global distribution of Methane in the atmosphere greatly increased.The Japanese Greenhouse Gas Satellite (GOSAT) is the only satellite that measures the column mixing ratio of atmospheric Methane. Since the 1990, various global models have been used to simulate CH4 concentrations. High-resolution simulation of CH4 at hourly intervals on Earth, with diverse ecosystems, due to the lack of intensive spatial and temporal measurements and the impossibility of reliable validations for chemical simulations are known as a serious challenge. The main purpose of this study is to understand the performance of the WRF-GHG model in simulation of Methane concentration and validation the results of medium-scale modeling output in total Methane concentration in comparison with GOSAT satellite observations over Iran.

Iran and some area of its surroundings is considered as a study area. This study focuses on two case periods of hot and dry (August 31-2010) and cold and wet (February 1-28, 2010). In order to provide the initial and boundary conditions of meteorological fields, ERA5 reanalysis data were used with a horizontal resolution of 0.25 ° with a time resolution of 6 hours. Different emission input data from three different global greenhouse gas emission databases EDGAR_v5.0 (anthropogenic emissions), GFAS emissions (fire emissions), and datasets (CMS_V01) (wetland emissions) have been used.Preliminary and boundary conditions for the chemical fields taken from atmospheric monitoring service data (CAMS) with a spatial resolution of 0.8º with 137 vertical levels and with 6 hours time resolution. To investigate and quantify the validity of meteorological fields simulated by the WRF-Chem numerical model, a set of observations of selective synoptic stations is used. For validation of CH4 WRF_Chem column concentration and statistical analysis, in the points that include remote sensing data (GOSAT sensor data), is used the set of level 2 products generated by the NIES algorithm. The local transit time of the GOSAT satellite Is approximated around 13: 00_9: 00, so the simulated concentration for this time is applied in the analysis. The first 15 days of the simulation are omitted to take into account the spin-up time. Statistical parameters of mean bias error (MBE), mean absolute error value (MAE), root mean square error (RMSE), and Pearson correlation coefficient (R) in meteorological and chemical variables are studied for validation of numerical simulations and quantification of error levels.

The model has been able to calculate temporal changes in surface temperature, relative humidity, and wind speed to some extent correctly. The general tendency of the model to simulate the observed temperature and relative humidity for the selected time period is evaluated. In general, model values are closer to summer observations than all thirty days in two selective months. The wind speed forecast is often consistent with the wind speed values obtained from the measurements, and in most cases, the wind speed is overestimated at around 1.2 m/s.Statistical evaluations of the WRF-Chem model, together with the GHG gas-phase chemistry mechanism, show the simulation of Methane concentration versus observations by the GOSAT satellite, and the estimation of the average monthly concentration in February and August 2010. The values of MAE, RMSE, RMSE_u, RMSE_s, BIAS and R are calculated equal to 42.92, 46.05, 7.82, 44.60, -24.99 and 0.63, for hot and equal to 12.01, 13.94, 7.09, 11.68, 7.50 and 0.76 ppb, for cold periods respectively. It can be seen that the WRF-Chem model performed better in Methane simulation in cold and wet periods (January) compared to the hot and dry seasons (August).

In this study, the WRF-Chem model was used to simulate meteorological variables and air pollutants (methane greenhouse gas) concentrations in the Middle East-Iran region during the study period of February and August 2010. The sensitivity of the model is considered using the GHG gas-phase chemistry scheme. The main findings of this study are: The model is able to reproduce temporal changes in surface temperature, relative humidity, and wind. The model underestimates the air temperature and relative humidity respectively around, 1/05 ⸰C -5% in the study area (Iran).In simulating of Methane concentration, and examining the results with related GOSAT satellite observations, the model overestimates around15 ppb of the Methane concentrations. The evaluation results show that the WRF-Chem model performs better in the cold season (January) than in the warm season (August). This uncertainty in CH4 simulation can be attributed to a deficiency in various input components of the CH4 emission in different categories. Improving the simulation for the various parameters reported to the model as the primary CH4 emission can generally help to improve the CH4 simulations.

This climatehascharacteristics and behaviors that distinguish it from the surrounding areas.Gradually, living and inanimateelements adapt to those conditions.As a result, any sudden change beyond the normal range and its normalbehaviorcauses theclimateof biologicalelements tobe stressedand stressed. These abnormalities maybe caused by climatechangeorlocal or human factors.In this study, we have tried to first examine the fluctuationsor changesin the temperature of the region in each of its parts in the last three decades.Temperature anomalies were alsoassessed for sea level in the same three decades. Synoptic patterns lead to anomalies in the temperature (hot and cold advection) are in cold season. In this study, atmospheric data of 31 stations from northwestern to western synoptic stations of Iran were used. To create a database with the same statistical database, statistical periods from 1989 to 2018 are arranged. To detect climate change, the software for detecting climate change, provided by the (Meteorological Organization), has been used. The statistical period studied is divided into three ten-year statistical periods and the changes in these three statistical periods were calculated for all cold months. In another part of the work for the colder months of the year in three decades, the status of sea level temperature anomalies was also examined using NCEP / NCAR site maps.The standardized Z index was used to identify the predominant synoptic pattern.The final selectioncriteria anomalies station high frequencyof 50% of the stations has been studied.So if the abnormalities are more frequent in the 15 stationandwas elected as a representative on the end of 202 days in the cold period has had this feature. Also,sea level data were extracted in the same range for 202 days during a script.The data of sea level equationwas converted to numerical data by scripting in Grads software environment, amatrix with dimensions of 203* 253 was analyzed by rotation with array in SPSS environment by factor analysis method, and by quarry ax method.Eleven factorsjustify 95.40percent of climate behavior in the cold months. Finally, after examining 11 factors, 4 dominant patterns were identified in this region.The results of this study showed that the temperature difference in northwestern Iran with southwesternand Middle Western is decreasing. The temperature difference betweenthe southandthe north in thefirst decade (1989-1989) was between 7 and 9 degreesCelsius, dependingon the month.in the second decade (1999-2008)this difference decreased to 5.5to 6.5 degreesCelsiusand in thethird decade(2009-2018) it decreased to 5 to 6 degrees Celsius.This means that the temperaturerangebetween the north and south of the country is declining.This means that the temperature range between the north and south of the country is declining. Among the colder months,January had the highest temperature fluctuations and April had the lowest temperature fluctuations.If we exclude January, in other months, a total of three decades in the northern part of the study area, especially in Ardabil province, the trend of increasing temperature changes and vice versa in the southern part of this increasing trend has had a slight slope. But in January, the phenomenon fluctuated sharply. In the second decade, compared to the first decade, there was an increase in both the northern and southern regions, while in the third decade, compared to the second decade, there was a decrease in both regions. Temperature anomaly maps prepared for the colder months of the year showed that from January 4 to April in thefirst decade we saw negative anomaliesthroughout the region.So in the first decade, compared to the previous decade, we have faced a negative anomaly throughout the region. But in the second and third decades, anomalies have been positive inalmost the entire region. This phenomenon has accelerated in the third decade.In November and December, the anomalies fluctuated sharply.In the first decade, the northwest was negative in both months, but the south had a positive anomaly.In the second decade, the situation was completely reversed and the northwestern region was associated with a positive anomaly and the southern part with a negative anomaly. But in the third decade, anomalies in almost the entire region tended to be positive and harmonized with other months.The results of visual examination and factor analysis on abnormal temperature days showed that four consecutive patterns of temperature abnormalities in the western and northwestern regions of Iran during the cold period. Therefore, four patterns are the cause of temperature changes and the general temperature of the region is outof its normal conditions in the cold period. As can be seen in the two models,the air temperature in the region is warmer than usual during this season of the year. This means that the temperature in the region is rising unusually and the general trend of the region is disrupted.In both cases, in addition to warming or rising temperatures, precipitation has also occurred. It is emphasized that not every precipitation system is associated with a warming. As can be seen, in both models, along with the warm rise of Saudi anticyclone, it plays a key role in hot advection. Two consecutive patterns have also been associated with cold weather, causing temperatures in the region to drop significantly. So the abnormal days that accompanied the cold. Or the temperature in the area has dropped unusually. Siberian high-pressure systems and Saudi and immigrant anticyclones have played a major role. In other words, the combined pattern of these three anticyclone systems has played a major role in cold advection and air stability in the region. In general, the temperature in the region is rising, and this upward trend is more intense in the northwest than in the south.The main systems that create abnormal days in the high-pressure Siberian and immigrant areas for cold advection and the Saudianticyclonehaveplayed arole in perpetuating these cold eruptions.Inother words,thecold weather has been advection to the region by two Siberian and migratory systems, and the anticyclone of Saudi Arabia in the higher layers has caused the reliability of this cold wave.On hot days, during the cold period of the year, Saudi Arabia's cyclone and Sudan's low pressure haveplayed a majorroleinhotadvection

Thermal temperature events for a period of 20 years (2001-2020) were extracted based on the data of synoptic meteorological stations in Mashhad, Sabzevar, Quchan, Neishabour, Bojnourd and Gonbad Kavous and the atmospheric patterns leading to these events were investigated. In this period, a total of 158 cases of cold waves and 1,014 cases of heat waves have been recorded in selected stations in the study area.From a statistical point of view, after using the Sen’s slope estimator nonparametric method, it has been shown that there was no significant increase in the intensity of cold waves for none of the stations and the highest trend of decreasing the intensity of cold waves is related to Mashhad station by 0.045 degrees Celsius per year. The number of cold waves during the statistical period has increased in Quchan, Neishabour and Gonbad Kavous stations and has decreased in Mashhad, Sabzevar and Bojnourd stations. The highest decreasing trend of heat waves is related to Quchan station by 0.056 degrees Celsius per year and no sharp increasing trend has been observed for any of the stations and at any of the significant levels. The number of heat waves during the statistical period has increased in Sabzevar, Neishabour and Gonbad Kavous stations and has decreased in Mashhad, Quchan and Bojnourd stations. From the point of view of synoptic analysis, strengthening of two high pressure systems (Siberia and Zinc in Europe) and expansion of orbital tabs of these two systems and creating high pressure belt in latitudes about 35 to 55 degrees north and altitude banding pattern of 500 hPa Northwest of the Aral Sea and the impact of the study area on the active tilt canal can lead to a noticeable decrease in temperature and the occurrence of a cold wave, which is associated with northeast winds of 850 hPa. In contrast, the spread of the European high-pressure system to the northeast of the country, along with the region's impact from the high-altitude subtropical ridge centered on North Africa and increasing the geopotential height of 500 hPa compared to the long-term average of the same period causes heat waves and temperature increases.Thermal temperature events for a period of 20 years (2001-2020) were extracted based on the data of synoptic meteorological stations in Mashhad, Sabzevar, Quchan, Neishabour, Bojnourd and Gonbad Kavous and the atmospheric patterns leading to these events were investigated. In this period, a total of 158 cases of cold waves and 1,014 cases of heat waves have been recorded in selected stations in the study area.From a statistical point of view, after using the Sen’s slope estimator nonparametric method, it has been shown that there was no significant increase in the intensity of cold waves for none of the stations and the highest trend of decreasing the intensity of cold waves is related to Mashhad station by 0.045 degrees Celsius per year. The number of cold waves during the statistical period has increased in Quchan, Neishabour and Gonbad Kavous stations and has decreased in Mashhad, Sabzevar and Bojnourd stations. The highest decreasing trend of heat waves is related to Quchan station by 0.056 degrees Celsius per year and no sharp increasing trend has been observed for any of the stations and at any of the significant levels. The number of heat waves during the statistical period has increased in Sabzevar, Neishabour and Gonbad Kavous stations and has decreased in Mashhad, Quchan and Bojnourd stations. From the point of view of synoptic analysis, strengthening of two high pressure systems (Siberia and Zinc in Europe) and expansion of orbital tabs of these two systems and creating high pressure belt in latitudes about 35 to 55 degrees north and altitude banding pattern of 500 hPa Northwest of the Aral Sea and the impact of the study area on the active tilt canal can lead to a noticeable decrease in temperature and the occurrence of a cold wave, which is associated with northeast winds of 850 hPa. In contrast, the spread of the European high-pressure system to the northeast of the country, along with the region's impact from the high-altitude subtropical ridge centered on North Africa and increasing the geopotential height of 500 hPa compared to the long-term average of the same period causes heat waves and temperature increases.

Climate change impacts on climate variables are among the challenges of large cities. In this regard, General Circulation Models (GCMs) are among the most reliable tools for assessment of the future climate variables. In 5th assessment report on climate change, the Intergovernmental Panel on Climate Change (IPCC) has applied the Coupled Model Intercomparison Project, Phase 5 (CMIP5) models. These models use scenarios called Representative Concentration Pathway (RCP). In current study, the assessment of the climate variables of Tehran, Iran, under climate change impacts was addressed. The preceding studies for Tehran show they need to be updated with newer scenarios and more downscaling models. Furthermore, there is merit in using more synoptic stations due to the vastness of Tehran. Given these findings, the current study focuses on two main objectives. The first objective is to assess the climate variables under the RCP scenarios in Tehran for 2021-2040. To this end, the eight CMIP5 models under RCP2.6, RCP4.5 and RCP8.5 were used. Accordingly, seven climate variables including mean Temperature (Tmean), maximum Temperature (Tmax), minimum Temperature (Tmin), precipitation, relative humidity, mean Wind speed (Wmean) and the sunshine hours were used and simulated for baseline period (1989-2018) and then assessed for future period. In the Second objective, for downscaling the CMIP5s, in addition to use of the Statistical DownScaling Model (SDSM), Fuzzy logic was also applied for downscaling. Accordingly, the Fuzzy DownScaling Model (FDSM) was generated and the performances of FDSM and SDSM were analyzed.

In this study, to assess the Tehran climate variables under RCP scenarios, the multi-model ensemble were applied to reduce the CMIP5’s uncertainties. Accordingly, the eight CMIP5s including CanESM2, CNRM-CM5, CSIRO-Mk3.6, FGOALS-g2, GFDL-CM3, HadGEM2-ES, MIROC-ESM-CHEM and MPI-ESM-MR were used. Given the uncertainty caused by the different outputs of the eight CMIP5s, the weighted means of the models’ outputs were used to calculate the daily climate variables for future (according to the ability of the models in simulating the baseline period). For this purpose, first, the CMIP5s were downscaled. In this context, the SDSM software (version 5.3.5) was used and also FDSM was generated. Then the performances of FDSM and SDSM were analyzed. On this basis, the superior downscaling models were selected using the comparison of simulation results and the statistical indicators of R2, RMSE, NSE and MAE. Accordingly, the CMIP5’s outputs were downscaled using the superior downscaling models and then the daily values of each climate variable were calculated. In the calibration and validation of the downscaling models at baseline period, the predictors were selected from the daily data of the National Center for Environmental Prediction (NCEP) using correlation test in SDSM software. Furthermore, in developing the FDSM, the Fuzzy C-Means Clustering process was applied, to determine the Fuzzy Membership Functions and the relevant Fuzzy Rules. By using the structure obtained by clustering, the FDSM was built as a Mamdani Fuzzy Inference System. In this context, the FDSM was developed in MATLAB software using the trial and error process.

By correlation test in SDSM software, the predictors were selected for the SDSM and FDSM models. Accordingly, the SDSM and FDSM were developed using the daily climate variable and the selected predictors. The performance analysis of both downscaling models (based on the statistical indicators of R2, RMSE, NSE and MAE and the comparison of simulation results in baseline period) demonstrate very good quality and performance for all the daily Tehran climate variables. Therefore, the Fuzzy approach has an appropriate capability in simulating and downscaling the climate variables. In addition, neither model has absolute superiority over the other in downscaling. However, it appears that with a slight margin, the FDSM had a better performance for Tmean, Tmax and Tmin, and SDSM had a better performance for precipitation, relative humidity, Wmean and the sunshine. Accordingly these models were chosen as the superior downscaling models. The results of future period show the increasing trend of annual changes in Tmean and Tmax, precipitation and the Wmean. The maximum increase of annual average in Tmean and Tmax and the Wmean among all scenarios will be in the order of 1.29oC, 1.57oC and 0.8m/s (for RCP8.5) and also the maximum increases of annual average precipitation will be 10mm (for RCP2.6). Furthermore, the month long-term averages of Tmean and Tmax in all three scenarios show significant increases in summer. For precipitation, relative stability in summer, and increases in winter and early spring are projected, but the changes in Tmin, relative humidity and sunshine indicate relative stability.

In this study, two main objectives including the assessment of the climate variables under the RCP scenarios in Tehran for 2021-2040 and also the performance analysis of Fuzzy logic in downscaling were addressed. The performance analysis of FDSM and SDSM demonstrated the high performance of both models and the appropriate ability of the Fuzzy approach in downscaling the Tehran climate variables. Therefore, taking the Fuzzy approach for downscaling has a technical justification. In this context, the application of the Mamdani Fuzzy Inference System and the Fuzzy C-Means Clustering increases the accuracy and quality of the results at different conditions. According to the results, the annual changes in Tmean, Tmax and the Wmean at all the three RCP scenarios will have an increasing trend, while precipitation will also (marginally) increase. However, the other variables will have a relative stability. From the point of view of monthly changes, there were noticeable increases in the long-term means of Tmean and Tmax in the future period during the months of July, August and September (i.e. summer season). As regards to precipitation, a relatively stable trend was observed in comparison with the baseline during the warm months of future period, but during winter and in particular at the beginning of the spring of the 2021–2040 period, there will be more precipitation at different months of the year than the 1989–2018 period.

One of custom methods for forecasting climatic variables is time series approach. We used this method for long term precipitation in Mashhad synoptic station. The annual precipitation series has three components: trend, cyclic variations and random fluctuations. The two components of trend and cyclic are not showed in statistical period of less than 100 years in arid and semi-arid regions .Therefore, linear, nonlinear, heuristic, meta-heuristic methods of time series patterns cannot explain this phenomenon well. In this study, the annual and long-term precipitation time series of Mashhad station with a statistical duration of 125 years was considered. The trend in data, cyclic variation are considered. Box and Jenkins’s method (1976) for time series is fitted on data and optimal selection of parameters and best performance are studied. Also, test for outlier in data and diagnostic analysis for residual are done.

In this research, the modeling of annual and long-term precipitation time series of Mashhad synoptic station with a statistical duration of 125 years (1894-2018) was investigated. At first, the trend of data in mean is tested using Mann-Kendall and Sen Approaches. Then, the variance of data was de-trended by box-cox transformation. Cyclic changes were considered by fitting polynomials from six to 12 degree. Optimal selection of the number of pattern parameters was based on Autocorrelation function (ACF), Partial Autocorrelation function (PACF), Extended Autocorrelation function (EACF) and Akaike (AIC) and Bayesian Criterion (BIC) criterions. Because ACF and PACF are good for distinguishing ARI (p, d) and IMA (d, q) patterns and if we have ARIMA(p, d, q) pattern we should use EACF. The patterns performance was evaluated by criterions such as the Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), Mean Percentage Error (MPE), Normalized Mean Squared Error (NMSE), Signed Mean Squared Error (SMSE) and U.Tow type of outlier test, Additive Outlier and Innovational Outlier are done on all observation. Also, Diagnostic analysis is done on residuals to consider their behavior over time, normality and independent.

First, the significant trend was evaluated by using the non-parametric Man-Kendall and Sen Tests at a significant level of 95%. The results showed that the annual precipitation in Mashhad does not have a significant trend in mean. But, Data has trend in variance which stabilized by box-cox transformation. Cyclic changes by fitting polynomials from six to 12 degrees displayed that none of them were significant. But they can approximately show the wet and drought cycle in 1984-2018 years. Optimal selection of the number of pattern parameters was based on Autocorrelation function (ACF), Partial Autocorrelation function (PACF), Extended Autocorrelation function (EACF). Also, Akaike (AIC) and Bayesian Criterion (BIC) criterions are used. But result showed that IMA (1, 1), ARIMA (1,2,3) and IMA(1,3) have significant parameters. The patterns performance was evaluated by criterions such as the Mean Absolute Error (MAE), Mean Percentage Error (MPE), Normalized Mean Sauared Error(NMSE), Signed Mean Squared Error(SMSE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE) and U. The results presented that the IMA (1, 1) model has the optimal number of parameter in the model, significant parameter and the best performance. The observations do not have AO and IO outlier. The results of diagnostic analysis also demonstrated that residuals are stable over time, follow normal distribution and are independent. Therefore, the long-term annual precipitation sequence of Mashhad follows the white noise pattern and the best prediction of the amount of precipitation is the average of data.

The annual and long-term precipitation time series of Mashhad station with a statistical duration of 125 years was modeled by time series pattern. We found that data doesn’t have trend in mean but has trend in variance. There are not significant cyclic changes but using the long term data we can see wet and drought cycles better. IMA (1, 1), ARIMA (1,2,3) and IMA(1,3) have significant parameters. IMA (1, 1) model has the optimal number of parameter in the model, significant parameter and the best performance. The observations do not have any outlier and residuals are stable over time, follow normal distribution and are independent. The long-term annual precipitation time series of Mashhad synoptic station follows the white noise pattern.

In this study, the temperature data of central Iran stations in different months and hot, cold and passing seasons were investigated from a statistical point of view. The results showed that the presence of an upward trend in the cold season in eleven stations was confirmed and rejected in two stations. The highest slope of trend line in this season is estimated in Qom, Daran, Arak and Natanz stations with values of +0.53 and +0.52 degrees per 10 years. The lowest amount in this season was allocated to Isfahan station with a value of +0.17 degrees per ten years. In the passing season, the existence of an upward trend in one station has been confirmed and rejected in 11 stations. The highest slope of trend line in this season was estimated in Qom station with values of +0.40 degrees per 10 years. In the warm season, except in Yazd, Isfahan, Garmsar, Bafgh, Robat posht badam, Kashan, Khor va byabank, Naeen, Shahreza, Ardestan, Golpaygan and Kabotar abad stations, which is the slope of the descending line, in the rest of the stations, the uptrend has been confirmed. An upward trend in Shahrod, Ghom, Natanz and Daran stations has been confirmed and rejected in Arak and Semnan stations that the highest slope of trend line belongs to Dararn and Gom stations with values of +0.30 and +0.27 degrees per ten years. It seems that the existence of confirmed upward trend in the warm and transit seasons caused by climate change and the disturbance of environmental conditions governing the stations, causes in the decades ahead in the residential areas of Qom, Daran, Shahroud and Natanz, the conditions of life comfort to be out of balance and its residents in the use of conventional cooling ventilation such as water coolers to use Cooling ventilation such as gas chlorine, which imposes a high cost on households for energy consumption, should be turned on.

In last years, different methods for a post-processing the model outputs have been developed which provide a practical tool that combines the observed data and predictions of the model using an algorithm to reduce the systematic errors of the direct model outputs without the need for long historical data archives. It is well known that numerical weather prediction (NWP) models usually exhibit systematic errors in the forecasts of certain meteorological parameters especially near the surface (Galanis et al. , 2006). Direct numerical weather prediction model forecasts of near surface parameters often suffer from systematic errors mainly due to the low resolution of the model topography and inaccuracies in the physical parameterize schemes incorporated in the model. In recent years, the increasing demand for accurate weather forecasts has led to a steady improvement of the skill of numerical weather predictions at both global and regional scales. Despite these improvements, such predictions are still affected by imperfect initial conditions, numerical approximations, and simplification of the physical and chemical processes that govern the evolution of the atmosphere. These imperfections, approximations, and simplifications result in random and systematic errors (e.g., bias) that affect the predictions’ accuracy. Bias here is defined as the ‘‘difference of the central location of the forecasts and the observations” (Monache et al., 2011). In order to reduce the influence of the above mentioned drawbacks in the final output of a NWP model, a variety of approaches based on statistical methods has been used. Most of them are derived from Model Output Statistics (MOS), which are able to account for local effects and seasonal changes. One of the most successful approaches to this problem is the use of Kalman filters (Kalman, 1960; Kalman and Bucy, 1961; Galanis and Anadranistakis, 2002). They consist of a set of mathematical equations that provides an efficient computational solution of the least square method. This paper is organized as follows. In Section 2, we introduce different post-processing method. In Section 3, we show that how the filters are applied on the model outputs for average temperature at Mehrabad meteorological station. In Section 4, statistical results are presented and finally the paper is concluded in section .

In this paper three simple algorithms based on Fourier transform, artificial neural network and Kalman filter have been implemented to correct the average temperature Weather Research and Forecasting (WRF) model forecasts in Tehran Mehrabad station. The WRF forecasting model is a mesoscale atmospheric research as well as environmental forecasts. The first post-processing approach implemented in this study is Fourier transform (FT) which decomposes a function of time (a signal) into its constituent frequencies. The term Fourier transform refers to both the frequency domain representation and the mathematical operation that associates the frequency domain representation to a function of time. The Fourier transform of a function of time is itself a complex-valued function of frequency, whose magnitude (modulus) represents the amount of that frequency present in the original function, and whose argument is the phase offset of the basic sinusoid in that frequency(Bracewell, 1986).Second approach is performed in this study is Artificial Neural Network (ANN) techniques to correct daily average temperature (Hansen and Salamon, 1990) .Generally speaking, data driven processes are governed by systems of linear or non-linear equations, which describe the relationship between the WRF model and observation values of the system's output and the values of inputs. Last approach in this method referred to Kalman Filter. Kalman filtering, also known as linear quadratic estimation (LQE), is an algorithm that uses a series of measurements observed over time, containing statistical noise and other inaccuracies, and produces estimates of unknown variables that tend to be more accurate than those based on a single measurement alone, by estimating a joint probability distribution over the variables for each timeframe. The filter is named after Rudolf E. Kálmán, one of the primary developers of its theory. In practice, the Kalman filter is the statistically optimal sequential estimation procedure for dynamic systems. Any type of Kalman methods aims at eliminating the standard error and, thus, at vanishing the corresponding biases. The fulfillment of this requirement is the main criterion ensuring the credibility of the filter.

In this study, the performances of three post processing methods are tested with the WRF model runs .Statistical results show that the model error varies according to the season at the station and the model forecast is underestimated or overestimated on all days, meaning that the error is positive on all days or negative on all days; While this estimate becomes positive for some days and negative for others after the application of the Kalman refinement, this is evident in the significant reduction in the mean error that measures the bias. From the root mean square of the errors, it measures the scatter of the error, and although its reduction is noticeable after the application of the filter, but it is far from zero and indicates the existence of additional and deficient estimates.

In this paper three simple algorithms based on Fourier transform, artificial neural network and Kalman filter have been implemented to correct the average temperature model forecasts.Results show that all of different methods of predicting average temperature are presented, reduce numerical weather prediction’s systematic and random errors. One of the most successful approaches to this problem is use of Kalman filter. Among the methods, the Kalman refiner was able to modify the WRF model prediction to an acceptable level and greatly reduce the error rate by as much as 90%. The main advantage of this statistical methodology is the easy adaptation to any alteration of the observations as well as the fact that it may utilize short series of back- ground information.

Climate change is one of the effective phenomena on the ecosystems and their resources management. Today, water resources management has special aspects with climate change. Predicting water quality has the same important as water quantity. Increasing global temperature due to climate change has changed the precipitation and river flow-rate pattern in the watersheds. Hence, the quantity and quality of the water resources (especially surface water) is particularly affected by climate variations. Many studies have investigated the climate change impacts on weather parameters; however, changing in water quality factors caused by climate variations has been less studied. Hence, this study has focused on the variations of weather parameters (temperature and precipitation) and forecasting of the surface water quality under different future climate scenarios. For this purpose, the GCM (general circulation model) HadCM3 and LARS-WG down scaling models were used in Eskandari Watershed of Isfahan.

In this study, the changes of climatic parameters and surface water quality factors in Eskandari Watershed of Isfahan (located in central Iran) were investigated in the base (BP) and future period (FP). The daily minimum (MinT) and maximum (MaxT) temperatures and precipitation data of Daran Station during 1993-2013 (BP) were simulated using HadCM3 general circulation and LARS-WG down scaling models. These parameters then predicted under A1B, A2 and B1 scenarios for the FP (2020-2030). Thereafter, the main quality factors of the surface water including electrical conductivity (EC), acidity (pH) and sodium absorption ratio (SAR) in the FP were estimated by substituting the climatic parameters in the multivariate empirical models derived from water quality and climatic factors in the BP. Then the suitability of water for agriculture and drinking was investigated using Schoeller and Wilcox diagram.

Results of the study showed that the MinT and MaxT of the study area will increase under three selected scenarios in the FP except for MaxT which will remain in B1 emission scenario. Precipitation will decrease in all studied scenarios specially in A1B and A2. Surface water quality had a close relationship with the variations of the temperature and precipitation in BP. There will be a greater increase in EC under A2 scenario. There will be the greatest increase of temperature and decrease in precipitation in this scenario as compared to the others which justifies the increase in water salinity. Whereas, pH of the surface water will decrease under three scenarios in the FP. Furthermore, SAR will enhance in all three scenarios which shows water quality degradation in the FP. Changing water type-facies from Ca2+-CO32- in BP to the Mg2+-CO32- in FP under A1B an A2 scenarios also indicates the deterioration of the surface water quality in Eskandari watershed. Although, the water quality of the watershed will degrade, but it will still satisfy the criteria for drinking (based on the Schoeller diagram) and agricultural (based on the Wilcox diagram) uses. It should not be overlooked that the land use type will have more effect on surface water quality rather than the variations of temperature and precipitation in the future. Hence, future anthropogenic activities make it difficult to interpret the effect of predicted climate variations on water resources quality (especially surface water resources). However, monitoring of the factors that affect quality of the water is the essential part of a water resources management plan in a watershed.

Results showed that the MinT in the study area will increase in the FB under three studied scenarios (A1B, A2 and B1); however, MaxT will also increase under A1B and A2 scenarios but it will nearly remain without change under B1 emission scenario. The annual precipitation has also a declining trend in the FP under three emission scenarios which has a negative impact on the surface water quality. Results also showed that EC will increase under three studied scenarios; whereas, pH will decrease with regard to the BP. Besides, SAR will increase as 17% in the FP as compared to the BP. The quality of surface water in the Eskandari watershed will satisfy the required criteria for drinking and irrigation according to the Schoeller and Wilcox classification, respectively, in the FP under three emission scenarios.

One of the main goals in climate studies is to do climate classification. Climatic divisions and the identification of the most important factors affecting each area is one way of identifying the climatic profile of the areas. The climate of each area is composed of all the factors and climatic elements of that area and all those factors and elements must be taken into account when segmenting. Only by referring to temperature, precipitation and humidity can one study the climate of an area (Khosravi and Aramesh, 2012). Climate classification has been the focus of climatologists since the early 20th century. So far, three types of climate classification (empirical classification, genetic classification and multivariate classification) have been applied. Today, the advent of computer technology and the advancement of statistics has provided the conditions for expanding climate information basics. Many climatologists have used this method to classify and map climatic elements. Carbajal et al. (2007) demonstrated the ability of factor analysis in zoning by zoning and zoning bioclimatic zones in central and northeastern Mexico. Carvalho et al. (2016) divided the European territory into regions with similar simulated climate change in a study using daily rainfall simulations and minimum and maximum temperatures. There have also been numerous studies on climate zone zoning in Iran. Masoudian (2003) has studied the geographical distribution of precipitation in the country using factor analysis method. Absolute Gratitude and Evening (2006) dealt with climatic analysis of Bushehr province using cluster analysis. .Salehi et al. (2017) In a study, the climatic zoning of Kohgiluyeh and Boyerahmad province was analyzed using factor-cluster analysis and concluded that Kohgiluyeh and Boyerahmad province climate has five main components and eight climatic components.

The study area includes East Azarbaijan, West Azarbaijan, Ardabil and Zanjan provinces. This part of the country occupies 118670 square kilometers of land area, which covers about 7.5% of the country. For this purpose, data on 29 climatic variables of 19 synoptic stations were obtained from the Meteorological Organization from 1980 to 2010 and used for climatic zoning of northwest of Iran. After verification and verification of the data as test run and t-student test in SPSS software matrix with 19 * 29 dimensions (19 represents the number of stations and 29 represents the number of climate variables) was prepared. Climatic classification of the Northwest region. Also, due to the differing scale of measurement data, standard score of data was used for analysis. Finally, using Varimax-era factor analysis method to reduce the data matrix dimensions and identify the main climatic components of the northwest region, and the cluster analysis method through the input to the climatic zones of the study area.

In factor analysis, it first standardized climatic data, then the corresponding analysis was performed using the Equamax rotation method and rotation method. Factor analysis using the basis components and varimax models showed that the 19 climatic elements of the Northwest region can be summarized into six factors with respect to their intrinsic correlation. After analyzing the matrices, climatic elements (factor loadings matrix) of 19 * 5 were obtained indicating that the climate of the Azerbaijan region is mainly the result of six factors, which together account for 92% of the region's climate behavior. These factors are: temperature factor, atmospheric humidity, precipitation, wind and visibility limit, thunderstorm, glacial.After identifying the important factors in determining the climate of Azerbaijan region, climatic zoning was carried out using climatic and scientific elements. This was accomplished by using different spatial methods of climatic categories that had previously been reduced by analyzing the principal components. Different tree diagrams were drawn using different clustering methods. The northwestern region was divided into eight climatic zones based on 29 climatic components using cluster analysis method.

In modern methods, climate classification is a process in which the statistical nature of climatic data largely determines the boundaries of climatic zones, not the individual taste of the researcher. These methods do not limit the number of elements that can contribute to the climatic zoning and therefore, this classification can identify climates in which the magnitude of the spatial differences of many climatic elements is taken into account (Masoudian, 2011). In this regard, applying modern statistical methods such as principal component analysis and cluster analysis to identify sub-climates of northwestern region, compared to traditional or classical methods such as Demarten, Coupon, Ivanov, Ambrose and ... Full advantage of modern statistical methods In identifying the distinct micro-zones they show climatic differences. In the Azarbaijan region, despite the existence of synoptic systems, the role of synoptic systems has been neglected due to various geographical factors such as altitude, altitude orientation, latitude, proximity to Caspian Sea and Lake Urmia. This has caused the destruction of several climates in the study area. In this study, after obtaining data from 19 synoptic stations and 29 climatic variables using factor analysis and cluster analysis and IDW mediation model, the climatic zoning of northwest region was performed. The results showed that the climate of Azerbaijan region is mainly caused by six factors: thermal, atmospheric moisture, precipitation, wind and visibility limit, thunderstorm and frost. These six factors account for about 92% of Azerbaijan's climate behavior. Subsequently, Euclidean interval cluster analysis was applied to six factors and the region was subdivided into eight semi-arid and windy regions, cold and moderately humid, cold and semi-arid, semi-cold and precipitation, semi-cold and relatively rainy. It was subdivided into warm, semi-arid, semi-cold and thunderstorms, and slightly warm and semi-arid. It should be noted that since in some parts of the study area there are synoptic stations for various reasons and no meteorological data is recorded, definitely with the establishment of long-term meteorological stations and the recording of long-term information, more accurate results on climate can be obtained. Reached areas.

Floods caused by rainstorm are one of the most important natural hazards in most watersheds of Iran and damage various sectors every year. Therefore, planning and flood control can reduce potential hazards. Flood control in flood source areas, is one measure to reduce floods and this is usually done most effectively through hydrological modeling. So far, different methods and models have been used to determine flood-prone areas, but the efficiency of these methods in different climatic and environmental conditions is unknown and requires further research. Therefore, the purpose of this study is to evaluate the efficiency of ModClark model in HEC-HMS software and determining flood source areas in Darungar dam watershed in Dargaz city of Khorasan Razavi province.

In this research, spatial distributed hydrological ModClark model in the HEC-HMS was used. For this purpose, rainfall and runoff data of regional stations, digital elevation model (DEM) of the watershed, land use map, vegetation, and hydrological soil group maps were obtained from the regional offices.Tthe model inputs were extracted with ArcGIS software and HEC-GeoHMS . In this research, gridded curve number method for rainfall loss, ModClark method for rainfall-runoff, recession method for base flow and Lag method used for flood routing. In the next step, sensitivity analysis, calibration and validation of the model were performed using six rainfall-runoff events. Then the design rainfall in the return period of 10, 25 and 50 years were entered into the model. Then the contribution of each sub watershed in flood hydrograph in watershed outlet was determined by using Unit Flood Response method and Successive Single Sub watershed Elimination. Finally priority of each sub watershed in the flood peak discharge and flood volume at the main outlet, were determined.

The results of sensitivity analysis for the selected parameters revealed that the parameters related to the curve number (curve number map and CNratio) had the highest sensitivity and the storage coefficient had the lowest sensitivity. Then the model was calibrated with sensitive parameters. The mean values of NSE, RMSE and PBIAS were 0.90, 0.27 and -3.36% for calibration events and 0.862, 0.35 and -7.71% for validation events, respectively. Thus, the efficiency of the model for flood prediction was confirmed. The prioritization results showed that for the return period of 25 years, sub watersheds of 53, 9 and 56 had the highest participation and sub watersheds of 44, 69 and 43 had the lowest participation in the outlet peak flow. In the 50 years return period, sub watersheds of 53, 66 and 37 had the highest participation and sub watersheds of 2, 43 and 44 had the lowest participation in the peak flow. Prioritization of sub watersheds based on flood volume criterion, shows more consistent results, so that the first three priorities in different return periods belong the sub watersheds of 53, 66 and 37. Despite some differences in prioritization by these two criteria, the spatial distribution of different degrees of flood risk is almost similar. Based on peak flow criterion, most of these sub watershed are located in the Center and based on volume criterion, they are mostly located in the Northwest of the watershed. Southern sub watershed in both criteria are classified as low-risk and safe sub watershed.

The overall study results for flood source areas, without considering the area of each sub watershed, indicated that in addition to the geological factors and vegetation, there is a direct relationship between the contribution to peak flood of sub watershed and their slope. The results of the sub watershed prioritization reveal that in the short return period, most sub watershed had little contribution in the outlet peak flow, but by increasing return period, sub watershed prioritization becomes more stable. Based on the results, it was concluded that size and location of the sub watersheds does not affect their contribution to flood peak and volume. The results of this study can be used in planning flood control operations in the study area. Since there are many methods to prioritize sub-basins in terms of flooding, it is recommended that prioritization be conducted with other methods and their results to introduce the best compare models.

In this study, the results of 24 and 48 hour predictions of mid-scale WRF model with nested slopes with 18 and 6 km separations (implemented in Lorestan meteorology) and with 27 and 9 km separations (implementation) At the Institute of Meteorology and Atmospheric Sciences), without schematic, for a period of 2 months from March 1, 2019 to the end of April 2019 and compared with precipitation observation data for 10 synoptic meteorological stations in Lorestan. For this purpose, 2 * 2 agreement table was used for verification. The results obtained from PC skill score showed that in a period of 24 hours, the ranges of 27, 18 and 9 km in more than 80% of cases were able to occur or not rain. At the provincial level, correctly predict that this score was a minimum of 6% for a range of 6 km. Also, studies performed for a period of 48 hours showed that all slopes in more than 77% of cases showed the accuracy of the occurrence or absence of precipitation. The results of validation in this study have been for days with precipitation and no specific threshold has been considered for precipitation values. It is suggested that in order to improve the relative weakness of the model, validation quantities for specific thresholds (light precipitation, Moderate rainfall and heavy rainfall). The results obtained from the PC skill score showed that in a period of 24 hours, the slopes of 27, 18 and 9 km in more than 80% of cases were able to accurately predict the occurrence or non-occurrence of rainfall in the province. The range of 6 km was minimal and amounted to 67%. Also, studies performed for a period of 48 hours showed that all slopes in more than 77% of cases showed the accuracy of the occurrence or absence of precipitation.* The average quantity of B slope for 4 slopes showed that the number of precipitation forecasts for a period of 24 hours in the surveyed slopes is between 1.39 to 1.49 percent higher than the cases in which precipitation occurred, which indicates that the previous precipitation was higher. The occurrence of precipitation is relative to the occurrence of precipitation. For prediction over a 48-hour period, this quantity is slightly improved and has less error.* The average quantity of TS in the forecast for a period of 24 hours is more than 66% for the ranges of 27, 18 and 9 km, which with the increase of the forecast period to 48 hours, this quantity has improved and in all 4 ranges to more than 72%. it is arrived.* The quantity of FAR in a 24-hour period for all 4 ranges varies from 29 to 38% on average, indicating that only 29 to 38% of the precipitation predictions have not been met. In a period of 48 hours, this quantity has also improved a bit and has reached 21 to 26. The results of quantifying the H collision rate over a 24-hour period indicate that the three slopes of 27, 18 and 9 km had a high ability to predict the occurrence of positive precipitation. Also, in a period of 48 hours, almost all 4 domains had high power.* Examination of the average quantity of F shows that in a 24-hour period for slopes 9, 18 and 27 between 29 to 33% of the cases where no precipitation has occurred, the model has erroneously predicted that this error is 47% for a range of 6 km. Is. In a period of 48 hours, the rate of this error for 4 domains has reached 35 to 50%, which indicates an increase in this type of error with increasing time interval.* Quantity of Pierce PSS skill score in 24-hour period shows 39 to 69% improvement for the studied slopes, among which the 9 and 27 km slopes have better performance. In a period of 48 hours, the value of this quantity has reached 48 to 61%, which, unlike the period of 24 hours, is the best value in the range of 6 km. In order to evaluate the model more accurately, further case studies are suggested in different seasons of the year.* Increase model execution time.* The model should be executed with different schemas.* The results of validation in this study have been for days with precipitation and no specific threshold has been considered for precipitation values. It is suggested that in order to improve the relative weakness of the model, validation quantities for specific thresholds (light precipitation) , Moderate rainfall and heavy rainfall).* The results of this research should be compared with GFS and ECMWF models.

Today, the world's population in 2018 is 7,632,819,325 billion. %54 of the world's population lives in urban areas and that trend is expected to increase to 66 percent by 2050. In other words, half of the world's population currently lives in cities, and %95 of population growth will occur in developing cities over the next 90 years. Until 1800, only %3 of the population lived in cities, and by 1900, less than %15 lived in cities. Among the problems of urban development is one of the most important risk of air pollution, which has greatly affected the lives of human societies and has caused a lot of costs in the field of health and urban management. According to the World Health Organization, 9 out of 10 people breathe polluted air, and it is estimated that 7 million people die each year as a result of environmental and household pollution. Iran in general and the metropolis of Tehran in particular are no exception to this rule, so that according to statistical data in 2017, 29,098 people died due to air pollution in Iran..The number of factors affecting air pollution based on the method and cases studied is 18 factors. These factors include, Natural factors, Fossil Fuels, Environmentally friendly technologies, Migration and population density, transportation and traffic, Planning and policy making, green space, Lack of proper budget allocation, Urban location and focus of administrative and commercial centers, Establishment of a factory in the west of Tehran, Improper expansion of the city and construction of urban highways, Weakness in waste management and landfill.

The philosophical foundations of research are interpretive. The research orientation is applied. In this research, the research approach is combined. The research method in this research is qualitative and quantitative .The research layer was in the form of field and library activities. The research strategy in this research is survey. The objectives of the research are descriptive and data collection procedures using questionnaires and documents. Also, data analysis tools in this research have been performed using Wizard scenario software.

For this purpose, first, the factors affecting air pollution in Tehran must be identified, which has been achieved in the previous stages. Also, the classification of factors into primary and secondary has been done by experts. The classification of primary and secondary factors is such that factors, natural factors, fossil fuels, environmentally friendly technologies, establishment of a factory in the west of Tehran, urban location and concentration of administrative and commercial centers and transportation and traffic are among the primary factors. And factors, migration and population density, expansion and growth of industry, home and industrial cooling and heating equipment, design and program and policy, cultural factors, high-rise construction, e-government, allocation of appropriate budget, uncontrolled expansion of the city and construction Urban highways, waste management and landfilling, agricultural and mining activities are among the secondary factors.In order to perform the analysis process, it is necessary to define different modes (descriptors) for each identified indicator or factor, because to work with Wizard scenario software, different modes of a factor must be specified, in order to be able to accurately analyze the previous conditions. After identifying its modes and descriptors, a cross-impact analysis questionnaire was designed, which was completed by 10 experts in air pollution. Each of them has a master's or doctoral degree in environmental and policy-making. Then, by taking the mode (frequency of the most selected number) from the answer sheet of the experts, the final numbers in the form related to the software are loaded. By placing the factors and descriptors in the Wizard scenario software as well as the descriptions related to the descriptors in the matrix in the data output software, the Wizard scenario software has provided 6,718,464 scenarios, of which only two The scenario is adaptable and exploitable.

The city of Tehran is facing the same problem. However, the issue that is of special importance in this regard should be considered from a trans-organizational point of view, and serious and fundamental attention should be paid to this institutionalized problem in Tehran, because the costs of preventing air pollution will be far less than treating and eradicating it. In fact, according to the results obtained from the Wizard scenario software, structurally, the most important primary factors affecting air pollution in Tehran, including urban location and focus of administrative and commercial centers, environmentally friendly technologies, natural factors, Fossil fuel is the plant's location in the west of Tehran and transportation and traffic. In this category, we have examined the primary impact mechanisms on the system. Of course, in addition to being effective on the system, each mechanism has a certain amount of passive on the system, and there is no absolute active or absolute passive. Among the primary active mechanisms of urban location and concentration of commercial administrative centers, it has the greatest active factor on the system, while on the least active on the system and the most passive factor is transportation and traffic. Also, the most important secondary factors affecting air pollution in Tehran in terms of structure, according to the analysis of the grade system provided by the software, can be design, program and policy, allocation of appropriate budget, migration and population density, e-government, Western factors, respectively. He pointed to the expansion and growth of industry, the uncontrolled expansion of Tehran and the construction of urban highways, high-rise construction, waste management and landfill management, green space, home cooling and heating equipment, and agricultural and mining activities. In fact, these secondary mechanisms represent a discussion of the direct effects on the system.

In today's world, rapid and sustainable development is the ultimate goal of all countries. The main constraint to sustainable development is the climatic conditions of countries, including rainfall, which is most correlated with natural disasters. One of the methods used for prediction is deep neural networks. The main purpose of this study is to increase the accuracy of hourly rainfall forecasting in Khorasan Razavi region using one of the deep neural network methods. In this study, we present a deep neural network architecture using the Ron Neuron-Based Random Neural Decomposition Stochastic (RSDSAE) method for short-term precipitation prediction. In order to improve the accuracy, rough neural networks (RNNs) contribute to the rainfall uncertainty and to improve the speed and accuracy, the combination of the sparse algorithm with the above model has been used and We also used the predicted output data output of the WRF model, and tests are calculated on rain data and by two criteria of RMSE, MAE and with five self-esteem, respectively, 0.7912 and 0.7662, and has been able to better performance than the model (RSDAE) of itself Show. Keywords: Deep Neural Network, Precipitation Prediction, stacked denoising sparse autoencoder.

Rainfall is a climatic factor that affects many human activities such as agriculture, construction, electricity generation, etc. Therefore, having a proper method for predicting rainfall can take preventive and mitigating measures for natural disasters. Over the past few years, deep neural networks (DNNs) have been used as a successful mechanism for solving complex problems in areas such as machine vision, image recognition, etc. (Hern´andez, et al., 2016). Deep neural network is a set of multi-layered architectures that have been trained using unsupervised algorithms and have challenges in training deep neural networks such as slow network training, overprocessing, and so on. Most existing neural network models have three drawbacks: (Khodayar, et al., 2017)1- Most of the architecture is shallow.2. Some methods require handmade engineering features that are tedious.3- Most methods do not have direct knowledge about rainfall uncertainty.In this study, to solve problems one and two, we combined the proposed method of stacked denoising sparse autoencoder (SDSAE) and to solve problem three, we combined the proposed method with Neuron Rough.2.

2-1. Data and area of study:In this study, the study area is Khorasan Razavi province and the output of networked data is the WRF model, which is a mid-scale regional forecasting model. We considered eight days of regional precipitation hourly data as target variable data for training and testing algorithms. We prepared the data used from the output of two regional forecast model implementations (mid-scale atmospheric forecast model) on 26 and 28 October 2018 for each implementation with 4 days forecast.2-2. Steps to perform the proposed method:The proposed method has five main steps:1) Receive WRF output prediction data,2) Apply noise to input data.3) Validation of deep network parameters.4) Unsupervised learning with RSDAE to build deep multilayer network.5) Learning by monitoring (fine tuning) by BP algorithm with SGD method2-3. Evaluation criteriaTo evaluate the proposed method SRDSAE and RSDAE, we use two criteria, mean squared error (RMSE), absolute mean error (MAE).

Predicting rainfall is one of the most important and fundamental challenges for researchers. Properly predicting rainfall improves the lives of the general public and even the proper planning of governments for the correct use of rainfall. Various algorithms and methods have been proposed for forecasting issues in recent years. Deep learning methods are one of the most popular methods for predictive problems. In this research, we have used the stacked denoising autoencoder method to predict rainfall and combine this method with neuron-rough and sparse algorithm to increase the accuracy of the forecast. In this research, we applied the proposed method on the output data of the WRF mid-scale forecast model in Khorasan Razavi province. The proposed RSDSAE(Rough Neuron Based stacked denoising sparse autoencoder) method and RSDAE (Rough Neuron Based stacked denoising autoencoder) method have been compared and analyzed for better comparison in precipitation forecasting based on two evaluation parameters RMSE and MAE. The proposed method has been able to reduce the forecast error by reducing the RSDAE network and subsequently reduce the RMSE and MAE criteria, or in other words, improve the rainfall forecast.