پژوهشنامه مدیریت حوزه آبخیز
پیاپی 20 (پاییز و زمستان 1398)

In this research, we used the support vector machine (SVM), support vector machine combine with wavelet transform (W-SVM), ARMAX and ARIMA models to predict the monthly values of precipitation. The study considers monthly time series data for precipitation stations located in Hamedan province during a 25-year period (1998-2016). The 25-year simulation period was divided into 17 years for training, 4 years for calibration and 4 years for validation. Statistical comparison of the results was done by using correlation coefficient (r), root mean square error (RMSE), and standard error (SE). Results showed that ARIMA, Support Vector Machines, ARMAX and support vector machine combine with wavelet transform were ranked first to forth, respectively. Furthermore, the support vector machine has fewer adjustable parameters than other models. So, the model is able to predict precipitation with greater ease and less time. For this reason, it is preferable to other methods.

Rainfall prognosis plays an important role in drought management and planning of drinking water and agricultural water resources. Also, Future policies can be tailored to optimize spending and maximum productivity. In this study, the effect of large-scale climatic signals on rainfall in Mazandaran province was investigated. The first, the effect of climatic signals on precipitation simultaneously and with delay was studied by statistical methods (Pearson correlation coefficient) and then, using the M5Tree model, monthly rainfall was compared with related indices. Generally, the correlation coefficient between signals and precipitation showed that correlation with delay was greater than the coincidence. The results of the correlation study between climate indices and monthly precipitation with a one-step delay showed that there was a significant correlation with, AMM, NINO1 + 2, NINO3, NINO4, TNA and WHWP indices and rainfall in the Babolsar station. Also, AMM, NINO1 + 2, NINO3, ONI, TNA and WHWP indices had a significant correlation with rainfall at the Gharakhil station. Based on the findings, correlation between climatic signals and rainfall in Noshahr station, was significantly different with AMM, NINO1 + 2, NINO3, NINO3.4, TNA and WHWP indices. The correlation between Ramsar precipitation and climatic signals showed a significant difference with AMM, NINO1 + 2, NINO3, TNA and WHWP indices. The results of the M5Tree model indicated that, generally, in one-step delay, the best simulation performance will occur. Also, comparing the simulation results with Taylor's diagram showed that at all stations, the one-step delay values were closer to the observations. The advantage of delayed prediction is that the rainfall situation can be pre-determined and used to manage the water resources of the watershed.

In a natural ecosystem, land exploitation and change in environmental conditions, particularly changes in vegetation and land use effects on hydrological responses such as floods and erosion and sediment and ultimately causing heavy social-economic losses. Then, prediction of effect of land use changes on flood situation in the future decades and sedimentation of watersheds will be the way to deal with this phenomenon. To simulate the hydrological behavior in future decades, HEC-HMS model was calibrated and validated for the previous period, with changes in curve number and percentage of impermeability (as a result of land use changes). Then, predicted hydrograph for coming decades (2020, 2025 and 2030) and last (2004) were compared. The results show that the peak discharge and flood volume in the 2020, 2025 and 2030, will be increased in compared to the decade of 2004 equal to (52.8 and 20/41), (70.29 and 32.51) and (87.78 and 38.05), respectively will increase. Mean peak in the 2020, 2025 and 2030 will be increase (due to land use change), 16, 25 and 38.35 percent, respectively. In the next step, sediment changes in the future was investigated using the sediment rating curve in decades of 2020, 2025 and 2030. The results showed that the amount of sediment for 2020 will be increase 12.01 % compare to 2004, for 2025 compare to 2020, 15.14 % and for 2030 compare to 2025, 36.64 %. According to the results, land use change in future leads to increase in discharge and sediment, which could affect the hydrological and morphological conditions of watershed.

The control of runoff and soil erosion is one of main issues around the world. Therefore, the application of soil amendments is essential in protect water and soil. On the other hand, increasing the sludge production from factories is one of environmental problems that there are methods for beneficial from sludge production including convert them to Biochar. Therefore, the present study was conducted with aim of investigating the effect of produced biochar from sludge of incubation factory with values of 0.4, 0.8 and 1.6 t ha-1 on the variables of time to runoff, runoff volume, runoff coefficient, soil loss and sediment concentration. The stages of this research were conducted at laboratory conditions for rainfall intensity of 50 mm h-1 and the slope of 20 percent with three replications in plot scale. The results showed that after application of biochar with rates of 0.4, 0.8 and 1.6 t ha-1, time to runoff toward control treatment increased with rates of 71.19, 91.58 and 150.54 percent, respectively, runoff volume toward control treatment decreased with rates of 21.00, 31.40 and 36.47 percent, respectively and also runoff coefficient toward control treatment decreased with rates of 67.57, 56.94 and 46.35 percent. Biochar caused the reduction of soil loss with rates of 71.04, 77.65 and 85.63 percent, respectively, and the sediment concentration with rates of 63.16, 67.21 and 76.17 percent respectively. The results of the statistical test showed that the effect of different values of biochar on the components of time to runoff, runoff volume, runoff coefficient, soil loss and sediment concentration were significant at the confidence level of 99%.

Identifying and determining the flood potential and prioritizing of the flood discharge in the sub-basins is very important in terms of watershed management, flood control and watershed management projects. In many watersheds, it is difficult to investigate and determine the impact of each of the factors and even upstream sub-basins on creating an outlet flood, especially in the sub-basins without hydrometric station. The purpose of this study was to investigate the effective parameters in the occurrence of flood in the sub-basins of Koshkabad watershed and compare the two methods of Factor analysis and rational method for flood priority. Koshkabad Basin is located in the central part of the Hamedan province and hydrologically is the main part of the Gharachay basin that it flows into the Qom lake. In this study, to determine the potential of flood sub-basins and their prioritization, First, extracted of climatic and physiographic features in sub-basins, then provided flood potential map in the sub-basins using factor analysis method. Also, using the Rational method for estimating run off coefficient, flood intensity in the sub-basins was determined by considering the effective factors in the runoff coefficient of the area including slope, landuse and soil condition and were prepared Potential map and prioritization of flood sub-basins. Comparison results of the two methods shows that the potential map of the factor analysis method is very identical to the map of the flood potential map in the Rationnal method. The final results of prioritization in both methods indicated that the Abaro, Yelfan, Mary anadg and Dareh Moradbik sub-basins with Severe and very heavy flood Are in the first to fourth priority and Siakamar, Qaraachy and Koshkabad sub-basins with a relatively low to low flood intensity are in priority from eighth to tenth priority.

The Landslide is one of the natural hazards that causes a lot of mortal and financial losses in high-altitude and seismic mountainous areas annually. Considering the importance of roads as a pillar of sustainable forest management in this study, we have tried to estimate the sensitivity of landslide occurrence in 8 forestry series with using frequency ratio approach. Initially, the slip position in the series occurred with ground scrolling by GPS, then the effective factors in the occurrence of landslides were determined for the study area. Raster maps of the effective factors were prepared and classified. Considering the distribution of landslides recorded in the field and the categories of effective factors, the frequency ratio for each class of each factor was calculated and their relationship with the occurrence of landslide was investigated. The results of the relationship between positioning and slipping showed that the highest amount of landslides occurred on the lower side of the road, and by increasing the age of the roads, the amount of landslides is reduced. The relationship between soil and slip occurred on the class with the code 2.2.3, which represents the soil texture of loamy, loamy clay and clay that the highest number of slides was observed. Elevation class 300 to 600 meter has the highest number of landslides. The relationship between the slope and slip showed that with increasing the slope up to 40%, the number of landslides was increased, but then the trend was decreasing. The relationship between geological and landslide showed that the highest number of landslides was observed in the class with L-PLL2 ​​code representing relatively high peaks with moderate slope, average soil depth, stability and permeability. The relationship between rainfall and landslide factors showed that the greatest slip is occurring between 600 and 800 mm. In the northern, northeastern and northwest directions, the highest slip occurred. The evaluation results of the model with the ROC method indicated that the area under curve (AUC) was 0.688 (68.8%).

Due to soil degradation and abundant of sediment production, the exact recognition of affecting factors on gullies erosion, and the impact of various factors on gully’s formation and development, require a more accurate study. In this regard, in the present study the length of gully, the area above headcut, headcut slope, percent of closed grain, bare soil, cover vegetation, litter, saturated electrical conductivity, pH, percent of organic matter, sodium absorption ratio, and percent of clay, silt and sand were measured in 30 gullies in 2018. Factor analysis with SAS software (version 9.1) was performed to determine the factors affecting the samples classification. Backward regression method in SPSS software (version 16) was used to determine the affecting factors on extension length of gullies, considering the length of gullies as dependent variable and other measured parameters as independent variables. The results of this study showed that the saturated extract electrical conductivity, the area above headcut and the percentage of bare soil factors play the most important role on the length extension of gullies in the Ghazeian watershed, located in the north of Fars province. Therefore, it is recommended to control erosion at headcut and through establishing the adaptable and resistance canopy copy cover, reduces the uncover area and increases the roughness coefficient. Also, it is essential to have a watershed in different part (climate) of the country having gully erosion to perform accurate research on soil particles, canopy cover, type and rate of rainfall, runoff height, shear stress, flow velocity, morphological characteristics, type of slope and its cover, mechanism of creation and expansion of gullies, and the gully erosion control methods.

The variable and complex nature of the sediment load of rivers has led that the estimation of sediment entering the reservoirs and the production of long term sediment, for determining the lifetime of the structures encounter with the problem. Application of sediment rating curves is one of the most common methods for estimating the suspended sediment load of rivers. Regardless of the accuracy of sediment measurement in sampling and measuring the concentration of samples and also measuring the corresponding discharge and with due attention to the spatial and temporal changes of suspended sediment in river management and reform operations, the providing of sediment rating curves is one of the most important resources of error for estimating sediment. Therefore, providing time backup curves for sediment rating curves is necessary for evaluation of accuracy and precision of sediment rating curves. So, in this research are presented time backup curves for sediment rating curves for investigating temporal variations of the sediment rating curve and are applied for considering slope changes of curve and annual changes of it too. Also, in this research compared the results of the sediment rating curve method with the results of some statistical methods for estimating the suspended sediment load in three hydrometric stations in Gamasiab watershed. The results of validations showed that transfer function model due to take into account time parameter with SSE 0.00003 ton/day, 3.24 ton/day, 5.92 ton/day, respectively for stations: Sang-Surakh, Vasaj and Marvil has high precision in the forecasting and moreover it presents confidently unbiased forecasts for the future.

The conversion of rainfall to runoff in basins includes nonlinear relations between the complex interactions of various hydrological processes. In this study, without considering of predetermined structure, relationship between input and output system was derived individually from the nature of the data recorded. Also, the phase difference occurred between rainfall and runoff signals using cross-wavelet transform analysis. Then phase difference diagram was plotted for the single and compound events of Lighvan basin. By applying these phases at the time of the rainfall signals, all errors resulting from considering average losses in basin was minimized that in this study was introduced as "minimum error time position"(METP). Also, discharge forecasting for basin was carried out by Kalman filter model and optimization in Lighvan basin state space, using calibration step events. By applying this phase difference to effective rainfall components, the error resulting from the considering of average infiltration losses and φ index decreases to the minimum. Also, using the linear programming optimization method, the unit hydrograph for Lighvan basin was used as Kalman filter measurement model. The results showed that by applying the phases difference between rainfall and runoff signals and integration with Kalman filtering and linear programming (KF-LP-CW), the corrected Nash-Sutcliff coefficient were obtained 0.94 in both Calibration and verification steps. These values were obtained 0.63 and 0.68 in calibration and validation steps respectively, as compared to the state that phase difference was not applied (KF-LP method). Therefore, in this study, significant improvement was observed with the application of phase differences in effective precipitation components.

Estimating the runoff in the basins lacking statistics is always considered by researchers And managers in planning, development and implementation of many projects of water Resource. One of the methods for estimating the runoff is to use geomorphology instantaneous unit hydrograph which estimates the hydrograph of flood based on quantitative geomorphology factors. In this study, first the quantitative geomorphology of the Nash and Rousseau methods including k (scale parameter), n (shape parameter) and speed (Meters per second) were obtained. In the next step, instantaneous unit hydrograph was obtained by GIUH method and ultimately flood hydrograph was created for 20 selected events from 1991 to 2007 in kasilian region and compared to the observational hydrographs. Through sensitivity analysis, infiltration Index (φ) had higher sensitivity analysis, infiltration Index (φ) had higher sensitivity and n (Parameter of shape) were less sensitive. Also, by comparing peak discharge values and computational hydrograph widthsin the state where the discharge of hydrographs is 50 and 75% of peak discharge,observational data revealed that the Rousseau method and lower error than the Nash method and Nash method had a higher accuracy than Rousseau in terms of the volume and time.

Varied flows such as flooding resulting from dam break can happen due to earthquakes, seepage, bombing, overtopping, mistake in design and project performance as well as causing financial damages and human losses. Unsteady flow during flooding events has a high impact on flow field pattern, sediment motion, ecology and also, distribution of contamination in the river. In this study, the impact of various slopes, sediment diameters and unsteady flows with different discharges on sediment transport were carried out at the flume with 12 m length, 0.5 m width and height. In this research, two type of uniform sediment particle with 10 and 15 mm in diameter were spread over the bed of above mentioned flume with slope of 0.01 and 0.02. To build a dam physical model, a mechanical lift-gate was used at 2 m from the upstream inlet of the flume. Potential hydrostatic initial conditions of the fluvial experimental model resulted from dam-break was based on using three water depths of 35, 20 and 12 cm at the gate upstream and one water depth of 3cm at the downstream of the gate. By considering the above mentioned procedures, fluvial flows were produced and have been used with discharge between 0.07 to 0.66 m3/s. The result showed that by increasing the sediment particle size 1.35 times the sediment transport have been reduced by 1/4 for 35 cm upstream water depth and also by increasing the slope by 2 times from 0.01 to 0.023 for sediment with 15 mm and with initial upstream water depth condition was 35, 20 and 12 cm the sediment transport have been increased 3.42, 1.83 and 3.38 times respectively. The above mentioned points show the significance of fluvial flows and their effects on sediment transport and therefore the morphological changes of rivers. Result has been shown that high sediment transport rate and bed evolution was resulting from positive wave head and after that, sediment transport rate have been intensity decreased.

The aim of this research is spatial Simulation and land subsidence susceptibility mapping using maximum entropy model in Jiroft and Anbarabad Townships. At first, land subsidence locations were recognized using extensive field surveys and subsequently the land subsidence distribution map was made in the geographic information system. Then, each of effective factors on land subsidence occurred in study area such as: percent slope, aspect direction, altitude classes, profile curvature, plan curvature, topographic wetness index (TWI), distance of drainage, litology units, pizometric data, land use, and normalized difference vegetation index (NDVI) digitized in GIS environment. Then, using frequency ratio (FR) method, the weight of the classess of each factor and was determined. Finally, land subsidence susceptibility map in the study area was prepared using the model maximum entropy. The results of validation of model using 30% of the unused points in the modeling process and according receiver operating characteristic (ROC) showed that the map of land subsidence susceptibility obtained from the maximum entropy had the high accuracy of ARC value of 0.859 (85.9%). Therefore, the zoning map can play a significant role in water resource management and identification of critical areas according to extracting groundwater table in the study area.

Groundwater resources are considered as a largest reservoir of available freshwater in the world. According to the importance and scarcity of this precious resource, studying the groundwater quality changes is very important. Declines in water level of Urmia lake has led to increase in the salinity of water in this lake. In this paper, water level reduction effects were studied on qualitative characteristics of groundwater in the eastern part of Urmia lake from 2001-2002 to 2011-2012. With using geostatistical techniques in GIS, TDS &SAR were obtained with Circular, spherical, exponential, and Gaussian semi-variogram models. In most of year Gaussian model with RMSE between 0.44 to 0.94 for TDS and exponential model with RMSE between 0.52 and 1.47 for SAR were selected as the premier models. Groundwater quality changes were evaluated using Co-Kriging interpolation method and qualitative zones were plotted in the eastern aquifer of Urmia lake. The results indicated that water level reduction of Urmia lake had negative effects on the decline of groundwater quality of the eastern plains of Urmia lake. Also, water quality at different points reduced so that TDS varied approximately between 10% and 45% and SAR varied between 7% and 30% during the statistical period.

Nitrate is one of the most important pollutants of surface water and underground water resources which has adverse effects on the health of consumers. Due to lack of sufficient opportunity and budget, in practice, full and timely sampling of all wells in each area is not possible. So the amount of nitrate in certain parts according to the adjacent groundwater is of particular importance in studies. In this study, data from 55 observation wells at 902 Km2 of Ghaemshahr-Jouyar plain were used. The accuracy of deterministic methods and geomorphology were evaluated to determine the most suitable method for determining spatial variations of nitrate. Among all the methods of interpolation, simple co-krigination with auxiliary calcium was identified as the most appropriate method. RMSE value was obtained 7.45 mg/l, MAE equal 5.95 mg/l and The coefficient of determination 0.47 . J-Bessel variogram model has the most powerful spatial structure. Using the land use map of Mazandaran province in the Research area also showed that the highest nitrate values were observed in the urban area. The results showed that the highest nitrate pollution was in Behnamir and from the coastal areas of the Caspian Sea and in all areas it is within the permitted limits.

One of the most important types of water erosion, is massive erosion. Soils with fine particles (especially silt particles) have little stability against erosion, because they have more water holding capacity and thus increase the saturation moisture and the likelihood of this type of erosion. So In this study, to investigate the relationship between particle size of silt, with behavior and rheological properties of soil in the rabor region of Kerman province, three profiles were studied respectively, at the top of the slope, middle slope and down hill slope. After identifying the soil layers, soil sampling was conducted at various depths. Then some physical and chemical characteristics of the soil, Includes equivalent calcium carbonate, sand, clay, silt, silt components, saturated moisture content and psychological limits was measured in triplicates. The rheological tests such as yield stress and viscosity were also performed to determine the flow behavior of the soil. The results showed that in the study area are high erodibility index position of the bottom and top of hill slope (k=0/54 down the slope and k=0/52 above the slope. The soil profile at the top of hill slope had the highest amount of coarse (28.33%) and saturation moisture (57.29%) and the lowest amount of clay (16.45%). Subsequently it had shear stress much lower than the position of the tilt tolerance and thus faster flow decreases. The shear stress was much lower than the position of the tilt tolerance. The best model to predict the flow behavior of soil was herschel-bulkley model, which has a higher coefficient of determination (R2>0.90) and lower SD than the Bingham and Kasson models. The results also showed that all samples have thixotropic behavior. The fine silt and calcium carbonate equivalent affected behavior and thixotropic properties. The fine silt mostly influnced the shear stress while the calcium carbonate equivalent affected the recovery process structure, viscosity and yield stress and yield stress was a good criterion for predicting soil behavior against sulification erosion and identifying the critical points of this erosion.

The intensive use of groundwater resources has often affected ground water levels in many parts of the world. Therefore, the strategic management of water resources for better planning of water resources seems necessary, In this study SWOT strategy was applied for strategic management of water resources in Abarkouh plain. In order to launch this strategy, we first extracted the weaknesses, strengths, opportunities and threats using group decision making, and then the relative importance and overall value of each factor assessment were determined. The process involved identifying weaknesses and strengths of the study area and opportunities and threats present in the water resources. These parameters were extracted using collaborative decision-making and then the relative importance and final value of each factors in the strategic SWOT analysis matrix were determined. The results of this study showed that the external and internal factors evaluation matrix was 2.25 and 2.18, respectively, which indicates overcoming the weakness to power and threats to opportunities. In other words, the condition of the Abarkuh Plain is located in the Weaknesses-Threats (WT) column with a defensive strategy. These results indicate that water resources condition is critical in Abarkuh Plain. To manage this crisis, the best management strategies were extracted from the strategic SWOT analysis matrix including: agricultural well equipment using flow meters, changing irrigation methods (underground, injecting, dripping) changing the crop patterns (pistachio, pomegranate and other low-water products), using non-living windbreaker and greenhouse shading, and finally drinking and sanitation water separating.

In the coastal region reduction of groundwater is one of the most important problems. Excessive exploitation from groundwater is one of the main effective reasons that will lead to the progress of saline water, especially in seasons with low precipitation. Therefore, at the current research, identification of suitable areas for artificial groundwater recharge in the Tajan plain (with an area of approximately 33467 hectares), was conducted. For this purpose, the necessary maps including land use, geology, gradient, soil texture, dry soil thickness (non-saturated soil part) and distance from the waterway were prepared.  Then the final weight of criteria and sub-criteria were determined by using AHP method; After that, by using GIS software, the mentioned maps were combined and finally the appropriate areas were identified and prioritized. Then the different methods of artificial recharging in study area were evaluated. According to the results, 6.91, 38.80, 25.15, 21 and 8.14 percent of total studied area were completely appropriate, appropriate, normal, inappropriate and completely inappropriate for artificial recharging respectively. Because of the short distance between study area and coastal region, some options like flood spreading and construction of rubber dam is suggested to artificially recharge of groundwater and to prevent of the saline water's advance.

There is a multifaceted approach of water resource in a watershed, which is accompanied by complexity and issues such as the interaction of natural and human systems. Achieving the consistent governance of water resources is one of the most important management approaches in the water sector, which requires a coherent structure of knowledge transfer and information on the organizations stakeholders’ network associated with water resources. In this research has been studied in Taleghan watershed to measure cooperation relations (transfer of water resources management knowledge), between 28 organizational practitioners (Including development, protective, and mediation organizations). In this study data were collected based on the questionnaire analysis questionnaire. The results show that the knowledge structure of water resources management in Taleghan watershed is highly integrated among organizational stakeholders and the type of knowledge management is currently not based on negotiation. On this basis, it can be claimed that based on the bonding social capital index (relationships of each of the sub- groups) is modest and based on the E - I index, bridging relations (relations between different subgroups) have a weak rate. it can be concluded that the relationship between water resource management and water resources management in Taleghan district level is an essential step to establish a cooperative governance of water resources.

This study aims to evaluate two methods of surface water routing including kinematic wave and time-area in upstream of the Bustan dam. To this end, first considering the features and application prerequisites of both methods and describing the hydraulic and hydrologic conditions within the Tamer watershed, algorithms and scripts of kinematic wave and time-area methods were developed. Then the scripts were run to simulate flow hydrograph at outlet of the watershed. The results of performance evaluation showed that Nash-Sutcliffe coefficient for kinematic wave method is 0.882 and 0.745, respectively for the calibration and validation data sets, while for the time – area method it is 0.903 and 0.748, respectively. These results showed a difference in timing of simulated and observed hydrographs especially in time to peak of hydrographs. One of the reasons for this non-compliance in hydrograph timing can be attributed to the lack of enough rain gauge stations across the watershed and errors associated with spatial distribution of rainfall on rain input maps. According to rainfall zoning map using GIS, two rainfall polygons were specified across the watershed and rainfall in each polygon was considered as uniform, but there might be a significant difference in rainfall pattern within each polygons which can be a main source of errors in timing of simulated hydrographs. In general, according to the assumptions and limitations of both models and also likely errors in rain input maps of events, performance of these surface runoff routing methods are considered suitable for simulation of outflow hydrographs. A main limitation of kinematic wave is assumption that the hydraulic gradient is equal to the bed slope which limits its application only for small watersheds with higher slopes. In the distributed time-area method, the storage effect of pixels is very small and negligible, whereas the storage effects of whole watershed is remarkably high depending on drainage density and watershed shape factor.

In the arid and semi-arid area's rainfall have considerable changes in terms of time and amount that make the water resource management an important issue. In this research, using both Boolean and Fuzzy logic were zoned potentiality suitable areas for construction of underground dams. The study area is located in the central region of Ardabil province with 7461 km2 area and semi-arid climate. For this purpose, were used many criteria consist of topography, geology, land use, geomorphology, water quality and quantity, depth of alluvial, and channels ranking. In the Boolean logic method, after preparing the related criteria, suitable and unsuitable areas were determined in two classes 0 and 1 that were combined using AND operator. In the Fuzzy logic method, the criteria were prepared in many classes and scored from 0 to 1 regarding previous researches and area conditions, then the criteria were combined with Fuzzy gamma operator. There is a good accordance between used criteria and combined maps that obtained from two methods. Overall central area and part of southwest and northern area determined as suitable locations for construction of underground dams. Results of Fuzzy logic showed that in the study area, the streams with 3 rank and deep alluvial areas have a high potentiality for underground dams’ construction. Therefore, these areas should be considered as the suitable areas for underground dams.

Mehrgerd region of Semirom county has faced the water shortage for several years. The emergence of drought has led to a decline in water resources in recent years; and there are various reasons for climate change and increased consumption in order to reduce water resources. The present study aimed to investigate the process of temperature and precipitation variations and its possible effects on groundwater resources. Therefore, the Mann–Kendall test was applied to the climatic data of five selected stations during a 28-year statistical period as well as the groundwater level statistics of 33 observation wells. Effects of temporal temperature and precipitation variations on water resources were assessed using the correlation between climatic parameters, SPI drought index and water resources statistics. Results indicated that the average monthly temperature had an ascending trend in Mehrgerd region especially in winter and it could lead to changes in types of precipitation and early snow melting. The precipitation had no significant trend in most months (except for the observed precipitation reduction in January and March). Groundwater resources had a significant descending trend in almost all months of year. The study on relationships of precipitation parameter and water resources of region indicated that the precipitation parameter affected groundwater resources with a delay of 6 months. The effects of wet and dry periods on the status of groundwater level are quite obvious; and the groundwater level of plain has had an annual decline since the onset of a drought period in 2008. Despite the fact that this level has fluctuated with the severity and weakness of drought in subsequent years, but the reduction of water level has been the overall trend of groundwater level in Mehrgerd plain.

Drought is one of the most important threats for agricultural production in the world, espatially in Iran. This phenomenon has started with reduction of precipitations that lead to a decrease in soil moisture and an increase in soil surface temperature, which affect the vegetation growth. The purpose of this study was to investigate droughts in the Jaz_Murian basin by VHI, VCI, TCI and SPI indices using MODIS satellite imagery and meteorological data. For this purpose, the LST products (MOD11A1) and vegetation products (MOD13A1) were selected from 2000 to 2018 at the same time (the first week of March). The VHI and VCI maps were produced and the drought occurrences with critical and very severe classes were identified. Also, the SPI index was calculated using precipitation data in Kahnooj synoptic meteorological station. According to the VHI index, the results showed that over 75% of the study area was classified as the severe, very severe and extreme classes of drought and has suffered from drought in 2001, 2002, 2006, 2009, 2012, 2016 and 2018. Very severe and extreme classes are covered the center of the region, and the eastern and southeastern parts of the basin that likely is related to the rainfall reduction and topographic characteristics of the area and poor vegetation. In comparison with the SPI meteorological index in the short periods (one month, three months, six months), the VHI index indicates the more correlation with the six-month SPI. According to the results obtained in this study, a large part of the study area is located in severe and very severe of drought classes which need more attention to water resource management in this region.

The accurate determination of river flow in watersheds without sufficient data is one of the major challenges in hydrology. In this regard, given the diversity of existing hydrological models, selection of an appropriate model requires evaluation of the performance of the hydrological models in each region. The objective of this study was to compare the performance of artificial neural network (ANN) and IHACRES integrated model to predict the flow of sivand river in Tashak Bakhtegan watershed located in Fars province as a warm and arid area. Calibration and validation procedures were done by using data from 1982-1995 and 1996- 2012, respectively. Neural Network Toolbox of MATLAB software were used to evaluate the capabilities of neural networks. In both calibration and validation periods, simulated flows by the IHACRES model for flood flows, were less than the observed data. The determination coefficients of the model during calibration and validation were 0.62 and 0.54, respectively. The determination coefficients of dynamic neural networks and static neural networks during calibration and validation ranged from 0.88- 0.94 and 0.51- 0.69, respectively. The results demonstrated that artificial neural networks predicted monthly flow of sivand river more accurately than the IHACRES model.