جستجوی مقالات مرتبط با کلیدواژه « Sensitivity Analysis » در نشریات گروه « آبخیزداری، بیابان، محیط زیست، مرتع »

The study of erosion and sedimentation is a crucial aspect and an integral part of watershed studies in a basin. The methods employed to prevent erosion and sedimentation play a significant role in basin management planning. However, due to the lack of accurate statistics on erosion and sediment amounts, it becomes necessary to utilize estimation models for assessing erosion and sedimentation. These models are often region-specific and may have a wide range of uncertainty in different areas. One notable benefit of modeling, regardless of its type, is its ability to analyze natural processes within the basin. On the other hand, understanding erosion and sedimentation processes in a region can be an effective step in managing, planning, and prioritizing available resources. In this research, the aim was to determine the sensitivity of the MPSIAC model to the relationships among its parameters. Additionally, the goal was to assess the model's sensitivity to changes in these parameters and their impact on sedimentation. By identifying the influential ground and atmospheric parameters, it becomes possible to prioritize their effect on water sedimentation in the region.

Through the sensitivity analysis of the model, it is possible to identify the flexibility of the model to the changes of its various parameters. It is also possible to determine the relationships between model variables and prioritize the parameters affecting the model output. Sensitivity analysis can be used in the calibration stage, in such a way that results are accurate, and time and cost are saved. In this study, to analyze the sensitivity of the MPSIAC model, first, the nine parameters of the model were examined. Then, according to the arrangement of different parameters in the structure of this model, the parameters were numbered from X1 to X9. Then, the score range of each parameter was entered in its calculation table according to the modified model of PSIAC. In this step, the average of each parameter was calculated according to its upper and lower limits. To achieve this, the nine parameters of the model were standardized, and the sedimentation rate was estimated by varying each parameter within its lower and upper limits while keeping the other parameter(s) fixed at their averages. With each change, the sediment discharge was calculated.

In the analysis of the results, the slope of the curves of the standardized parameters relative to the dependent variable (specific sediment) is considered. So that, any part of the curve that has more slope changes means that the model is more sensitive to minor changes in that parameter in that interval. In this study, the results of this sensitivity analysis reveal that the slope factor and runoff volume have a considerable influence on sediment yield. For slopes up to approximately 15%, the slope parameter has a relatively smaller impact compared to other parameters. However, as the slope value increases, its effect becomes more pronounced, indicating that high slopes (greater than 15%) have the greatest impact on sedimentation. The runoff factor exhibits a sudden increase in sensitivity at higher flow rates.

In examining the effect of different parameters on the model performance, the parameters that had the largest slope of changes can be considered as the most effective parameters in sedimentation of the basin. In other words, the model demonstrates high sensitivity to slight changes in specific peak discharge values above 20 m³/s/km2. Therefore, implementing integrated (biomechanical) management measures to control slope steepness and reduce runoff volume can be an effective strategy for mitigating sedimentation. Therefore, based on the results, the role of land factors is very important, such as slope and then runoff, in the erodibility and sediment production in basins. Of course, it should be noted that climatic parameters directly and indirectly affect the amount of peak flood discharge and should not be neglected for their importance and influence in flooding and subsequently the erodibility of the watershed.

As one of the most significant and sensitive landforms to wind erosion, sand dunes vary in their extent of activities based on the influence exerted on them by various climatic and terrestrial factors.The study area comprises a Busher province’s county located in southern Iran, where sand dunes activity causes great problems for the local community. Therefore, the identification of active sand dunes by using wind regime of the region and the factors involved in their activity based on the Lancaster index, the aim of this research is.

First, the trends of climate factors, including precipitation, temperature, potential evapotranspiration, wind velocity, and local and non-local-originated dust were investigated at an annual scale using the data concerning the hourly-based wind velocity, weather codes, monthly precipitation, temperature, and potential evapotranspiration data collected from Dayer meteorological station over a 30-year period (1989–2018). Then, the sand dunes’ mobility status was examined by calculating the frequency of erosive wind and the aridity index over the whole study period using the Lancaster Index. Moreover, Sensitivity analysis was performed to predict the effect of meteorological parameters’ potential changes on sand dunes' mobility.

The study’s results indicated that according to the Lancaster index, sand dunes were fully active during the 30-year period. Furthermore, the aridity index was found to be subjected to real, extremely serious, and serious risks of desertification in three classes hyper-arid, arid, and semi-arid, respectively. On the other hand, the results of sensitivity analysis suggested that if the frequency of upper-erosive-threshold winds would increase by 30% in the future, the sand dunes' activity and mobility would increase by 30 %. However, if the precipitation rate increases by 30 %, sand dunes' mobility, and activity will decrease by 23%. Moreover, no significant trend was observed in meteorological parameters, including precipitation, temperature, and potential evapotranspiration throughout the whole study period.

In general, an increasing trend was found in sand dunes activity during the 30-year study period, suggesting a direct correlation between such an increase and precipitation. Moreover, the study’s results indicated that the management measures carried out over the past 30 years (for instance, planting seedlings, hand-planting afforestation, and grazing management) have proved insufficient, necessitating the application of new management policies in this regard.

Long-term monitoring as well as the impacts of climate change on sand dune mobility play a crucial role in reducing their adverse effects, including creating dusty days, potential risks of reduced air quality, and other environmental and health risks. The main objectives of this study were to investigate the dust phenomenon and its origin, to analyze the mobility status of sand dunes using two indicators, to determine the effect of climatic conditions on mobility, and also to predict the effects of changes in climate parameters on sand dunes mobility in the strategic area of ​​Sarakhs, located on the northeastern border of the country, for a period of 27 years. The results on the number of dusty days showed that the local dust events at Sarakhs station had a significant increasing trend. The results of sand dune mobility using Lancaster and modified Lancaster indices showed that there was a significant relationship at 99% level with dusty days, based on which the ability to use the modified Lancaster index in this area is raised. It can also be acknowledged that in the study region, increasing environmental drought coming with the mobility of sand dunes increases the risk of desertification. Also, the results of sand mobility sensitivity analysis showed that the combined effect of the frequency of erosive winds or the potential of sand transport and evapotranspiration has the greatest effect on reducing and increasing sand activity in the future.

Sand cat Felis Margarita (Felidae) is a small cat distributed in deserts of North Africa, Arabia, Central Asia, Pakistan and central Iran. This species is listed as Least Concern (LC) by IUCN and categorized on appendix II of the CITES. Successful conservation of this species demands for a comprehensive understanding of its habitat requirements. In this study, a Bayesian Belief Network (BBN) habitat suitability model was developed for sand cat through a practice of knowledge elicitation, calibration, sensitivity analysis and verification. Denning possibility, prey density, distance to Haloxylon woods, and soil stability were recognized as the most important variables determining habitat suitability of sand cat. Overall accuracy of the model was calculated at 96.7% using 120 presence/absence points collected across the species habitat in eastern Isfahan province. The results of model evaluation by sensitivity analysis and using data from fieldwork showed that the prepared model works well to determine habitat suitability. The results of this model not only combined the expert opinion and quantitative data obtained from land use maps, but also provided a good basis for use in adaptive management for the management of sand cat habitats.

Soil erosion is an important subject in water and soil conservation researches which is influenced by natural and anthropogenic factors. Hence, knowledge on soil erosion amount enables the identification of critical areas and prioritization of measures. One of the effective factors in the evaluation of soil erosion is slope and length (LS) which could be calculated using different methods and algorithms, so the selection of the best method to provide proper estimation is important. However, the comparison of the performance of the different estimation methods has not been sufficiently considered. Therefore, the present study was conducted to calculate the LS factor and evaluate its effect on estimations of soil erosion in the Shazand Watershed using four common algorithms viz. Renard et al. (1997); Desmet and Govers (1996); Moore et al., 1991, and Böhner and Selige (2006) in geographical information system. The results of this study while confirming the difference of about 15 times among performances of various algorithms, indicated that the mean soil erosion using the algorithm of Renard et al. (1997); Desmet and Govers (1996); Moore et al. (1991), and Böhner and Selige (2006) were 4.95, 19.47, 1.73, 1.34 t ha-1 yr-1 in case of the stability of other factors of RUSLE model. Considering the calculated amounts of slope length on the topographic map, the Desmet and Govers (1996) algorithm performed better than other algorithms in LS calculation for the study watershed. It clearly verified the necessity of selecting pertinent procedures for the calculation of input factors for the precise estimation of soil erosion.

 The accurate estimation of actual evapotranspiration (ETA), i.e. crop water requirements, is an important issue for irrigation water allocation at fields and improving water efficiency. In the present study, ETA for Sorghum crop was estimated using surface energy balance algorithm for land (SEBAL) and by employing Landsat8 satellite images; meteorological data and DEM map and then compared with the measured values of lysimeter.

 For carrying out the sensitivity analysis, the key parameters of SEBAL algorithm was altered at the range of ±10 , ±20 , ±30 , ±40  , ±50  and resultantly the variation of ETA corresponding to decrease and increase of these parameters was investigated. The changes of key parameters on the days of 206, 238 and 254, based on Julian days calendar, were evaluated at eight points with different vegetation densities. Albedo parameters have moderate to high sensitivity, so that the albedo parameter has medium and high sensitivity in the areas with ET> 10 and ET <10, respectively.

 The results of SEBAL showed the value of accuracy indices were acceptable at a significant level of 95%, while compared with those of lysimeter measurements. The results of the sensitivity analysis also showed the surface temperature and input shortwave radiation are more sensitive especially at the areas having low ETA values. The wind speed and

In general, in addition to high economic efficiency compared to other conventional methods, the SEBAL algorithm has a good efficiency in estimating ET. Also, performing sensitivity analysis and determining the key input parameters, while increasing the accuracy of measuring those parameters, can greatly improve the modeling results.

Sand dunes are one of the most dynamic geomorphic features of the earth's surface. Accordingly, identifying the factors affecting the mobility of sands and predicting their future status is essential to control dust. The study area was selected due to the location of Sabzevar city in the busiest railway line of the country, and the condition of sand dunes around this city. For this purpose, after calculating the percentage of erosive winds and drought index, the Lancaster index based on the average annual rainfall ratio and annual potential evapotranspiration for the period of 1990-2016 was calculated for analysis of sand dunes' mobility status. To predict the effect of climate change on the mobility of sand dunes, a sensitivity analysis test was carried out. The results show that the values ​​of the Lancaster index in this station did not show the inactive status in this span time period. The results of the analysis of the relationship between drought index and mobility of sand dunes show a significant effect of drought on the mobility and activity of sand dunes. Finally, the results of sensitivity analysis showed that if the frequency of erosive winds and evapotranspiration potential increases by 30%, the activity of sand dunes in the Sabzevar station will increase by 38%. Also, due to the potential of solid particles in creating dust, the location of sand dunes, and the study of local dust at Sabzevar station, the priority is to stabilize the hills located in the east of the study area.

Accurate estimation of reference crop evapotranspiration (ETo) is important in hydrological studies and water resource management that can influence the planning and policies for optimal allocation of agricultural water resources. For this purpose, in the present study, sensitivity analysis was employed to evaluate the effects of the changes in meteorological parameters (vapor pressure deficit (VPD), Mean Temperature (Tmean), net solar radiation (Rn), wind speed (u) and actual vapor pressure (ea)) on reference crop evapotranspiration (ETo) within the possible range of ±20% on a daily and growing season scale. ET0 was calculated by The Food and Agriculture Organization of the United Nations (FAO) 56 Penman-Monteith approach using long-term climate data from 13 meteorological stations of Khorasan Razavi province in northeast Iran. The sensitivity analysis indicated that ETo changes positively with vapor pressure deficit, mean temperature, net solar radiation and wind speed, while it changes negatively with actual vapor pressure. Also, over the whole period, ETo was most sensitive to vapor pressure deficit (VPD), followed by mean temperature (Tmean), net solar radiation (RN), actual vapor pressure (ea) and wind speed (WS). In the the warm growing season, ETo was most sensitive to vapor pressure deficit (VPD), followed by mean temperature (Tmean), net solar radiation (RN), wind speed (WS) and actual vapor pressure (ea) while, in the cool growing season, ETo was most sensitive to vapor pressure deficit (VPD), followed by actual vapor pressure (ea), Mean Temperature (Tmean), net solar radiation (RN) and wind speed (WS). The results showed that the sensitivity of ETo in different stations and different growth periods is different. The findings of this study showed that the importance of effective parameters on the reference crop evapotranspiration (ETo) was different between the hot and cool growing season of the region, which should be considered in sustainable planning of water resources and management of agriculture in Khorasan Razavi regions for development of adaptation strategies in arid climatic conditions.

Estimation of sediment bed load is one of the most important issues in river engineering. Accordingly, several methods and relations have been proposed to estimate the bed load in streams for the past decades. The present research aimed to select the most accurate method to estimate bed load through the testing of various relations to estimate the bed load of the Hablehroud River. The hydraulic and geometric specifications of the Hablehroud River at Bankouh hydrometry station were measured on specific dates by the expert team. The bed load rate was calculated by four different relations in Bankouh hydrometry station from 2011 to 2017. Resultant data showed that the Toffaletti method with RMSE, MAE, and R2 of 20.07, 10.86, and 83 estimates more accurate in comparison with other methods. The sensitivity analysis of the Toffaletti equation revealed that the bed load depends greatly on the velocity of the river flow. Such a method is firmly recommended to use in the mountainous region.

Although flood maps based on the deterministic approach play an important role in minimizing flood losses, there is considerable uncertainty in calculating the level of water inundation. Roughness is a key parameter in water surface elevation. Since roughness is not easily measurable and is estimated based on experimental and laboratory methods, it introduces a significant degree of uncertainty into the model. Therefore, in this paper, a probabilistic framework based on the Monte Carlo simulation approach is used to analyze the uncertainty of the roughness coefficient using the HEC-RAS model. In this framework, a computational module in Visual Basic language was introduced to communicate with the HECRASController to automatically run 84,000 Monte Carlo simulations within 7.5 km of the Karde River. In each performance, 1000 Monte Carlo simulations were performed in 84 separate cross-sections based on the Manning roughness probability distribution and the results of roughness uncertainty at water elevation were presented at 99%, 95%, 50%, 5%, and 1% confidence levels. The results indicate a significant effect of roughness on the water surface elevation uncertainty at 90% confidence level so that changes in water surface elevation in many cross-sections reach more than 1 meter. Comparing the cross-sections with the minimum and maximum uncertainty bands, show that the cross-sections with higher lateral gradients have more uncertainty and the V-shaped cross-sections with the lower lateral gradients have fewer uncertainty bands. Also, the results of sensitivity analysis to select the optimal number of simulations indicate that the bandwidth of uncertainty is not affected by the number of simulations of more than 1000 runs. The presented results based on the deterministic approach and increasing the reliability and improving the performance of the model based on the uncertainty of roughness in the Karde River can be used in the design of rainwater catchment.

The accurate estimation of daily Evapotranspiration (ET) improves the efficiency of water resources management especially in areas where suffers from water scarcity. In the present study, ET was estimated using surface energy balance algorithm for land (SEBAL) and the experimental model of FAO-Penman-Monteith (FPM) and finally compared and verified with those calculated from pan evaporation method. Since many climatic factors affect the ET values, the sensitivity analysis of SEBAL inputs variables was finally cerried out to determine the key affecting parameters. In this regard, by SEBAL model and emplying the satellite data of Landsat 8 (OLI and TIRS sensors), the ET values were estimated on a daily scale for the time period 2018/07/25 to 2018/09/11. Results of SEBAL model showed that the values of SEE, RMSE and R2 indices were equal to 1.27, 0.76 and 0.77 mm /day and 0.91, 0.6 and 0.92 mm /day, while compared with those of FPM and pan evaporation methods, respectively.

Over the past few decades due to population growth and urban development, urban runoff has increased and led to different problems such as inundation of urban pathways, dissemination of environmental pollutions and flood hazards. In order to urban runoff management, it is necessary to estimate runoff rate correctly. SWMM is one of the most widely used models in estimating urban runoff. The goal of this research is to evaluate the performance of SWMM model in simulating flow rate in an urban watershed in District 22 of Tehran. At first, model required parameters were calculated. For model evaluation and validation, in three events, runoff was measured in the watershed outlet and was compared with simulated runoff. The model validation results showed that the simulated flow rates had good adaptation with the observed ones. The validation results were used for estimating optimum values of model input parameters. The results of SWMM model evaluation confirm model accuracy with NS= 0.72 and RSR= 0.53 and indicate the model ability in simulating urban runoff. So, SWMM model can be used for urban runoff management plans and designing urban runoff drainage networks in this area.

Wind erosion is one of the important aspects of land degradation in arid and semi-arid areas. Countries located in arid and semiarid belt of the world including Iran have always been faced with this phenomenon. Wind erosion assessment models use different scores to determine the erosion rate in a given class. However, due to the spatial and temporal complexities and the multiplicity of factors affecting the ecological conditions of the region, it is impossible to fully rely on such results and use them for targeting and prioritizing the areas and providing suitable solutions for their management. Nevertheless, Bayesian Belief Networks (BBN) are based on probabilistic approaches which display the uncertainty in the evaluation of phenomena in terms of probability. These Networks are essentially developed as tools for analyzing decision-making strategies under uncertainty. Accordingly, this study set out to estimate the potential of the BBN as a relatively new and probable means for estimating the wind erosion and, finally, to evaluate the management scenarios for controlling wind erosion in Isfahan province.

To start the process of modeling the Bayesian Beliefs networks with regard to the purpose of the study, suitable variables were selected for modeling the BBN by reviewing the related literature and asking the experts opinions. In the next step, the relationships between the variables were determined, using the impact graph. The impact diagram shows the relationships and effects of the variables on each other and on the output node of the model (the amount of wind erosion). Finally, in order to create a model and formulate the conditional probability tables of model variables, the impact diagram was transformed into a BBN model, using the Netica software. The Bayesian Belief Network Model was validated by sensitivity analysis, the results of the analyses carried out by experts, and the comparison of the obtained map of Bayesian model with the wind erosion map in Isfahan province. The ROC curve characterizes the relative performance of each model. The area under this curve is called the AUC and the model with the highest AUC has the highest relative Function. The AUC equals 0.5 equivalent to the neutral model and the closer this value is to 1, the higher the Function of the model would be.

Having applied the final model of Bayesian’s belief network, the causal relationships between the variables affecting the rate of wind erosion were shown. The target variable in this model was wind erosion. As the findings of the study indicated, geological variables, land management, topography of the area, soil texture, rainfall and frequency of erosive winds were considered as key variables of the model. In order to run the model, information about each of the key variables was taken from the area at the province and fed into the model. Finally, the model was designed to estimate the amount of wind erosion in each sampling point. Based on the output of the model, the probability of wind erosion in each unit was used to zone the probability of wind erosion in the study area. The overall sensitivity analysis of the model also indicated that the wind erosion rate of the area had the most sensitivity to the wind velocity and speed, the frequency of erosive winds, and the protection of the earth's surface. On the other hand, the least sensitivity was reported for such variables as soil texture, geology and topography. Moreover, a high correlation between the results of the two models was found. According to the results of the ROC curve, the area under the AUC curve for Bayesian network model is 0.79, suggesting acceptable model accuracy.

It was shown that the BBN presented the probability of different wind erosion rates for each sampling point in the study area. In BBN, the results are expressed in the probability language and managers are to choose and implement timely and appropriate management decisions to reduce the risk of wind erosion in the region. The designed model in this study could be implemented in all regions. However, depending on the conditions of each region, the number of variables in the model may be increased or reduced. The study used Bayesian belief networks in the critical areas of wind erosion in Isfahan province. This model well demonstrates the importance of implementing wind erosion control projects to assist in sustainable land development to prevent migration, foster agricultural conservation, increase industry development and bio-resources in desert areas with desertification control.

Ephemeral Gully (EG) erosion is one of the most destructive types of water erosion, which contributes significantly to land degradation. EG erosion prediction is necessary to assess the magnitude of soil loss and to implement the appropriate conservation measures. The aim of current study  was to evaluate the efficiency of EGEM model for EG erosion prediction in Ghoorichay Watershed, Ardabil Province, Iran. For this purpose, a number of 17 EG erosion was identified and monitored between the years 2014 to 2016. The morphological characteristics and erosion rate of EGs were measured and recorded after seven effective rainfall events. In order to calculate the EG erosion, EGEM model requires four major categories of input data, including identification information, watershed data, soil data, and rainfall data. The model has two major components: hydrology and erosion. The runoff induced by a rainfall event in each gully catchment was determined by the Natural Recourses Conservation service (NRCS) Curve Number (CN). The result of EGEM model performance evaluation showed that the eroded soil volume and cross-section were predicted with a determination coefficient of 0.96 and 0.89, respectively. Sensitivity analysis revealed that the curve number was the most sensitive parameter, so that, with a 10% increasing and decreasing in CN, the volume of soil loss varied 22.98 and -18.92%, respectively. It can be concluded that EGEM model was suitable for event-based EG erosion prediction in Ghoorichay Watershed and it can be recommended for studying and planning on EGs in similar watersheds.

In the present study, the effect of climate change on the flow of the Hablehroud River was assessed, using the HadCM3 general circulation model, the LRAS-WG downscaling model, and the SWAT hydrologic model. First, using the LARS-WG model, the output of HadCM3 model in the period of 2011-2040 was downscaled and the necessary data for the SWAT model were obtained. The downscaling results showed that there is no meaningful difference between the future precipitation and the precipitation observed in the base period of 1995-2010. It was also observed that in the A2, B1, and A1B emission scenarios, the mean of minimum temperature will increase by 0.59, 0.47, and 0.43 0C, respectively, and the mean of maximum temperature will increase by 0.51, 0.39, and 0.35 0C. In the next stage, the SWAT model was calibrated and validated using the monthly flow data recorded at Bonkouh gaging station located at the outlet of the basin. The values of the coefficient of determination and Nash-Sutcliffe in validation phase were 0.8 and 0.72, respectively. Introducing the downscaled rainfall and temperature data to the SWAT model, the variation of the basin outflow was simulated in the period of 2011-2040. Results showed that in the three scenarios under study, the annual flow will increase 10.66, 17.72, and 17.52% in comparison to the base period.

Knowing the amount of runoff and sediment generated from different lands is an important step in land use management. Since it is not always possible to measure these values, modeling parameters, will be a way to achieve watershed comprehensive management. Beheshtabad watershed in Chaharmahal & Bakhtiari province with different land uses such as range land, agriculture, gardens and urban area is an important part of Northern Karun Basin. In order to simulate the amount of runoff and sediment generated in different land uses, SWAT model was used. Discharge and sediment data of weather and hydrometric stations located in the area was used for sensitivity analyses and then to calibrate and validate model by SUFI-2 algorithm. NS and R² coefficients obtained for runoff calibration are respectively 0.69, 0.71 and for runoff validation are 0.65, 0.67 and for sediment are 0.72, 0.73 in calibration, and 0.66, 0.71 in validation, that confirmed model ability to accurately estimate of flow, runoff and sediment in the study area. The results showed that dry farming lands with average runoff and sediment of 190 mm, 24.5 tons per hectare per year respectively, has the highest and gardens with an average of 80 mm runoff has the lowest sediment yield of 1.63 tons per hectare in the year.

Parks and green spaces are one of the most important elements of cities. The design and function of urban parks should be in line with the requirements of urban life and in response to the needs of citizens, as this can be used to create a healthy urban environment. The purpose of this research is to model the satisfaction of urban parks visitors using the artificial neural network. In this study, a multi-layer perceptron network was used to process the data with the intelligent neural network tool. First, 103 urban parks were selected in Karaj and Tehran, and information about regional, service and aesthetic variables was collected in all parks. Then, the collected data was considered as network input and the results of satisfaction level assessment as network output. The value of determination coefficient (R2) in this study was 0.72 which indicates the suitability of artificial neural network for satisfaction modeling in urban parks. The results of sensitivity analysis showed that variables of landscape quality, number of sports fields, food centers, and barbeque have had the most impact on satisfaction of urban parks. Therefore, in planning and managing public places such as urban green spaces, consideration of users' perceptions of the environment should be highlighted.

Spur dikes are one of the common methods to protect rivers against erosion. Scouring around the spur dike is an important factor that can disorder the structural performance. Using protective spur dike is proper technique reduce the scour amount. In this research, the GMDH and GEP model used in order to evaluate and estimate the effect of various parameters of protective spur dike on scour depth around the main spur dike. Important parameters consist of protective spur dike angle (Ө), protective spur dike length ( ), main spur dike length ( ), distance from main spur dike (X), flow intensity ( ) and Froude number ( ) are considered as the model inputs. Results of training set and testing set indicate that GMDH model is better than GEP model as the MAE and RMSE error in the testing set data are reduced from 0.063 and 0.086 (in GEP model) to 0.045 and 0.061 (in GMDH model) respectively. Also, the Nash-Sutcliffe criteria increased from 0.51 (in GEP model) to 0.75 (in GMDH model). In the following, using the GMDH and GEP models and according to the nature of the problems, the equations are suggested to predict the scour depth reduction in the first main spur dike. The results of sensitivity analysis indicate that the most effective parameter in decreasing the scour depth around the first main spur dike is ( ).