This study evaluates the performance of the linear first-order Volterra model for simulating nonlinear rainfall-runoff process. For this end, fifteen storm events over the Navrood River basin were collected. 70% and 30% of the events were used to calibrate and test the suitability of the model. Finally, the performance of the model was compared with the artificial neural networks (multilayer perceptron (MLP)) using five performance criteria namely; coefficient of efficiency, root mean square error, error of total volume, relative error of peak discharge and error of time for peak to arrive. Results indicated that the intelligent MLP models outperformed the Volterra model. The linear Volterra model was not more effective in simulating the rainfall-runoff process. It needs to be extended to higher orders and also the number of the parameters should be reduced.

Although many kinds of rainfall-runoff models have been developed by hydrologists, the unit hydrograph methods are still gainful tool for flood estimation in many basins where recorded hydrological data are not sufficient to support distributed rainfall- runoff models. The aim of this paper is to derive the analytical unit pulse response functions of quick and slow runoff of stream flow using a conceptual model containing three serial tanks as well as a parallel hybrid tank. The interrelation of the tanks can be shown using exponentially structured models which their parameters reflect the physiographical characteristics of the basin. Parameters of the model were estimated using Nelder–Mead optimization method. Efficiency and validity of the developed conceptual model were evaluated for various observed events using statistical criteria such as Nash-Sutcliffe, mean relative absolute error for discharge, peak flow and runoff depth. Results showed that the developed model had good capability in predicting rainfall- runoff process considering soil moisture conditions before rainfall occurrence without the need for definition of excess rainfall.

In the past decades, human activities changes the quality of water resources significantly so that water resources have been contaminated or to be contaminated in the future. In this way, agriculture has more contribution to reduce water quality and increase consumption of surface water than other activities. Agricultural management in river’ s watershed is one of the main approaches to improve the quality of water resources. Water quality simulation models in the basin scale, are the practical tools to simulate the impacts of various activities such as agriculture on the quality and quantity of water resources. In this study, the SWAT model to simulate the salinity load under two cycles, heavy metals and nitrogen in the catchment basin Navrood who represents the West Guilan, was used. The accuracy of the model to simulate the salinity in the two methods was compared. Calibration and validation of SWAT was conducted by data series of salinity and discharge during 2006-2013. To assessment the efficiency of model, two statistical indicators R2 and NS was calculated. The values of the indicators in simulating river flow were 0. 82 and 0. 58 respectively. In salinity simulation, these indicators were 0. 30 and-0. 71 under heavy metals cycle and 0. 61 and 0. 54 under nitrogen cycle, respectively. Therefore, the use of the nitrogen cycle in simulation of salinity load of Navrood basin is recommended.

In several types of rainfall-runoff models, the unit hydrograph based methods are useful tools for flood estimation in many, except non-gauged, basins. In this study, unit pulse response functions (quick and slow runoff) derived, considering linear system theory concept and using tank conceptual model. The model parameters were estimated with direct search optimization method. The model applicability and validity were verified using observed rainfall-runoff data of Navrood basin in Gillan province. The results showed that the tank model could simulate rainfall-runoff process with acceptable precision by taking into account the antecedent soil moisture conditions and without need to define excess rainfall.

Risk assessment of soil erosion, one of the most important land degradation problems worldwide, is very essential for land and water resources management, and development of soil conservation methods. In the present study, the temporal changes of soil erosion risk were assessed from 1987 to 2010, based on the Revised Universal Soil Loss Equation (RUSLE) using Remote Sensing (RS) and Geographic Information Systems (GIS) for the Navrood Watershed, Iran, with an area of 270 km2. Two Landsat satellite imageries obtained in 1987 and 2010 were used to assess the changes in vegetation cover during this period, and to obtain the Cover factor (C) of RUSLE. Rainfall and soil texture data and a digital elevation model were used to calculate the rest of RUSLE factors, which were taken as constant for the study period. The results showed that the average annual soil loss over the watershed ranged from 0 to 1, 056 t ha-1 y-1 (Cumulative percentage>99.9%). The area mapped as very high erosion risk (>100 t ha-1 y-1) increased from 10% in 1987 to 12% in 2010, and the area of the next risk class (51-100 t ha-1 y-1) increased from 8 to 9%. These changes cover an area of about 800 ha in the watershed, in which erosion risk has been doubled or tripled in the last 23 years. Forest clearing and rangeland overgrazing were identified as the most important reasons for the increase in erosion risk.

Background and Objectives: Climate Change can affect soil erosion, as the most important factor in the degradation of land in the world, by changing precipitation patterns. Therefore, it is essential to assess the impact of climate change on soil erosion risk. This study was aimed to evaluate the impacts of future climate change on soil erosion risk in Navrood watershed, located in west of Guilan province, North of Iran. Materials and Methods: In this study, the climate change trend was evaluated by XLSTAT software using effective climatic parameters obtained from Rasht and Bandar Anzali stations. Then, the soil erosion risk was predicted using RUSLE in combination with geographic information system and remote sensing, in Navrood watershed. The data of previous research were used to calculate the K, LS, C and P factors for the RUSLE model. The atmospheric general circulation models (NCCCSM), was used to produce synthetic weather series, over three A1B, A2 and B1 scenarios. Based on the outputs of NCCCSM, daily rainfall values of the base period 2002-2007 and the LARS-WG model, daily rainfall pattern were simulated for two 20-year periods of 2046-2065 and 2080-2099 for Kharajgil, Khalian and NAV stations located inside the watershed. Results: The results showed that the precipitation in the future will decrease in two stations of Khalian and Nav and will increase at the Kharjgil station. In contrast, due to increase of rainfall intensity, in all scenarios and stations the rainfall erosivity in the future is more than the base period. Based on the obtained results, soil erosion risk varies from zero to more than 77 tons per hectare per year, from zero to more than 115 tons per hectare per year and from zero to more than 98 ton per hectare per year for the base period (2002-2007) and 2046-2065 and 2080-2099 periods, respectively. Conclusion: The results showed that rainfall erosivity in the coming periods increases due to increasing rainfall intensity. Most area of the watershed has a low erosion risk and the southwest areas of the region and northern parts of the north are mainly at risk of erosion. Additionally, although rainfall erosivity is at its highest level at some parts, the amount of erosion is not high, which can be due to the effect of vegetation. Increasing vegetation density, particularly forest type, can reduce the effect of rainfall erosion and thus reduce the risk of erosion.

Basins in terms of geology, climate and other factors on the roughness and erosion are very different, the aim of this study was to evaluate Navrood watershed erosion, erosion rate, sediment yield and factors in the relationship between the physiographic features, topography, climate, geology, geomorphology, vegetation, soil erosion is generally determined. The basin has an area of about 265/46 square kilometers. The aim of this study was to evaluate the erosion and sediment production in the watershed and to find the sensitive areas of erosion. To achieve the above objective, the erosion potential (EPM) was used for data collection. The data collection tools, aerial photos, map types, methods and library sources are observed. In this study, using EPM (Erosion Potential Method) erosion rate in each sub-basin (sub-basin 11) has been investigated. In relation to the regional geomorphologic outcrop s is composed of 15 types. By integrating them into the structure of the slope, lithology and resistance to wear and combine them work units respectively. The erosion rate and sediment production in each experimental work units using EPM (quantitative) took place. Based on the results obtained, the following basins 4 and 5 of erosion is most severe erosion (erosion rate of 0.24 and 0.22). The highest sediment than the other sub-basins have been allocated. Because of the large number of specific erosion and deposition in the sub-basin, sub-basin compared to other high slopes and rock sensitivity to erosion, Switch, grassland and forest degradation, road-building.

Surface water management practices are directly influenced by the stream flow forecasting, especially for the low-flow context. In this paper, the monthly low flow time series were modeled and forecasted using a traditional stochastic model (Autoregressive Integrated Moving Average-ARIMA) and an artificial intelligence based model (Adaptive Network based Fuzzy Inference System-ANFIS). Low-flow in each month was defined as the minimum value of one, three, and seven day moving averages of daily stream flow. The performance of the stochastic model was compared to the neuro-fuzzy model through application to the stream flow data from the Navrood River basin in the Guilan state, northern Iran. The results showed that the stochastic model resulted in more accurate forecasted values than the neuro-fuzzy model for one, three, and seven day low-flow time series. Furthermore, in all neuro-fuzzy and stochastic models the error in forecasting three-day low-flow is less than those for one- and seven-day low flow.

Introduction: Water resources systems management is directly influenced by stream flow forecasting. It is therefore necessary to develop appropriate and applicable models for stream flow forecasting, especially in the low-flow context. Both stochastic models and artificial intelligence based models are widely used for simulation and forecasting of hydrologic time series. The literature shows that both models have performed well in different cases (Mishra et al., 2007) and thus it is difficult to know a priori which particular model would be better suited for a given problem.

Introduction: Due to the variation of rainfall and inflow regimes in different years, it is necessary to adopt a suitable management model for reservoir operation. Water requirement may differ at different demand sites and therefore it can be satisfied with one reservoir or a network of reservoirs. Different methodologies are adopted for optimal operation of a system of reservoirs. These may be classified in different groups, Stochastic Dynamic Programming (Braga et al., 1991, Ghahraman and Sepaskhah, 2002), Fuzzy Rule Based Programming (Faye et al., 1991), Linear Programming (Mohan and Raipure, 1992) Deterministic Dynamic Programming- Regression Based (Karamouz et al., 1992), Mixed Integer Non-Linear Programming (Teegavarapu and Simonovic, 2000), and Dynamic Programming And Artificial Neural Network (Chnddramouli and Raman, 2001),. Since nearly all dams in Iran regulate water for agricultural uses, a wise discipline should be adopted for best allocation of limited water (especially in droughts) considering the uncertain future river yield.