Developing an Interactive Spatial Multi-Attribute Decision Support System for Assessing Water Resources Allocation Scenarios

Water Resources Management - In water resource management, assessing water resource allocation scenarios (WRASs) is an important multi-attribute decision making (MADM) problem. It involves...     

A Comparative Assessment of Artificial Neural Network,Generalized Regression Neural Network,Least-Square Support Vector Regression,and K-Nearest Neighbor Regression for Monthly Streamflow Forecasting in Linear and Nonlinear Conditions

Monthly forecasting of streamflow is of particular importance in water resources management especially in the provision of rule curves for dams. In this paper, the performance of four data-driven models with different structures including Artificial Neural Network (ANN), Generalized Regression Neural Network (GRNN), Least Square-Support Vector Regression (LS-SVR), and K-Nearest Neighbor Regression (KNN) are evaluated in order to forecast monthly inflow to Karkheh dam, Iran, in linear and non-linear conditions while the optimized values of the model parameters are determined in the same condition via the Leave-One-Out Cross Validation (LOOCV) method. Results show that the performance of the models is different in linear and nonlinear conditions; the cumulative ranking of the models according to the three assessment criteria including NSE, RMSE and R² indicates that ANN performs best in linear conditions while LS-SVR, GRNN and KNN are in the next ranks, respectively. But in nonlinear conditions, the best performance belongs to LS-SVR, followed by KNN, ANN, and GRNN models.     

Water Management of Irrigation Dams Considering Climate Variation: Case Study of Zayandeh-rud Reservoir, Iran

Dependency of reservoir operation on the climate variation occurs especially in regions, where agricultural demand has a significant share of the total water demands. The variability between demands that are based on annual climate conditions may be larger than the uncertainty associated with other explanatory variables in long-term operation of an irrigation dam. This paper proposed a rule curves to the water managers of the Zayandeh-rud reservoir in Iran in long-lead reservoir operation. A regional optimal allocation of water among different crops and irrigation units is developed. The optimal allocation model is coupled with a reservoir operating model, which is developed based on the certain hedging that deals with the available water and the water demands mutually. This coupled model is able to activate restrictions on allocating water to agricultural demands considering variation of inflow to the reservoir, variation of demands and the economic value of allocating water among different crops and irrigation units. The resulted rule curve is presented with a number of tables for more details and accuracy and a simple curve, which is more useful for operational purpose.     

Development of a Hybrid Data Driven Model for Hydrological Estimation

High and low stremflow values forecasting is of great importance in field of water resources in order to mitigate the impacts of flood and drought. Most of water resources models deal with the problem of not being flexible for modeling maximum and minimum flows. To overcome that shortcoming, a combination of artificial neural network (ANN) models is developed in this study for monthly streamflow forecasting. A probabilistic neural network (PNN) is used to classify each of the input-output patterns and afterward, the classified data are forecasted using a modified multi-layer perceptron (MMLP). In addition, the performance of the MLP and generalized regression neural network (GRNN) in streamflow forecasting are investigated and compared to the proposed method. The findings indicate that the R² associated with the suggested model is 46 and 80% higher compared to MLP and GRNN models, respectively.     

Decision Support System for Monthly Operation of Hydropower Reservoirs: A Case Study

This paper presents a decision support system for multipurpose reservoir operation. The mathematical models in the system are formulated for monthly operation of hydropower reservoirs. The key components of the system are four main modules: database management, inflow modeling and forecasting, operation management, and real-time operation. Flexibility is the key feature of the system, providing the users with different decision tools and different indices for measuring the performance of each tool. A cost function is developed based on the present value of the total capital cost and the cost of operation and maintenance of the system. This cost function, which is developed based on “reasonable” estimates of water and energy prices, is used to measure the performance of reservoir operation policies. A utility function based on multicriterion decision making (MCDM) that uses an analytical hierarchy process is also developed. The MCDM utility function enables decision makers to incorporate the priority of different objectives in developing optimal operating policies and can be effectively used when the priority of objectives is not clear and the decision-making process relies mainly on the decision maker’s preferences. Both economic and MCDM utility functions are implemented and coupled with deterministic and stochastic optimization models. The decision support system (DSS) is applied to the largest surface water resources system in Iran, namely, the Dez and Karoon river-reservoir system. The results of the case study have shown that the DSS has been able to significantly increase the long-term power generation of the system while satisfying water demands for different purposes.     

An Approach for Probabilistic Hydrological Drought Forecasting

This paper proposes an approach to monitor and forecast hydrological drought in a probabilistic manner. The proposed approach deals with the supply and demand variables and the role of carryover in a system to estimate the probability of drought severity at different hydroclimatlogical conditions as well as different storage volume levels. This approach might be of significance when the supply and demand variables of a water resources system change considerably by climate variation. Major probability values and their mutual use in the proposed drought forecasting method are discussed. The presented approach is applied for the hydrological drought forecasting of Zayandeh-rud river basin in Iran. This probabilistic view of drought monitoring and forecasting is useful for risk-based decisions in water resources planning and management. The proposed index could be used to overcome the lack thereof in the existing surface water supply index.     

Developing a new method for spatial assessment of drought vulnerability (case study: Zayandeh‐Rood river basin in Iran)

The paper presents an approach to spatially representative depiction for assessing the vulnerability of central Iran's Zayandeh‐Rood river basin to drought using multiple indicators. Drought conditions prevailed in the study basin from 2002 to 2007, with an annual rainfall deficiency of 45 to 55%. Multi‐attribute decision making (MADM) methods develop a framework to evaluate the relative priorities of drought assessment based on a set of preferences, criteria and indicators. The proposed MADM process uses well‐known techniques for product weights analytic hierarchy process (AHP) and order preference (TOPSIS). These indicators include the Standard Precipitation Index (SPI), water demand, the Palmer Drought Severity Index (PDSI) and the Groundwater Balance and Surface Water Supply Index (SWSI). Indicators' spatial information was categorised in layers prepared in the spatial domain using a geographic information system (GIS). The alternatives were ranked and presented using TOPSIS. Results show that the proposed method was highly effective in representing assessments of drought vulnerability.     

Drought Mitigation through Long-Term Operation of Reservoirs: Case Study

Dealing with climate variability in a river basin presents many challenges in managing a water resources system. Occurrence of severe and persistent droughts deplete reservoirs storage to critical levels, which may lead to future water supply disaster. This paper illustrates certain benefits of using long-lead streamflow forecasts as well as restriction rules for reservoir operation to help manage the water resources system in the Zayandeh-rud River Basin in Iran. An approach is developed for activating restrictions on allocating water to agricultural demands during a drought and predicting low flow regimes using long-lead forecasts. The long-lead forecasts could utilize valuable hydroclimatic information such as the El-Nino southern Oscillation and northern Atlantic Oscillation to predict seasonal streamflow values. Hedging rules for optimal water supply releases is developed based on the benefit functions of release and carryover storage at each agricultural season. Hedging rules are triggered by different levels of drought indices determined by the predicted water availability at the beginning of each agricultural season. The method is used on an historical data set of hydroclimatic variables of the system to simulate the real time operation of the Zayandeh-rud Reservoir. The utility of the method is demonstrated for operating the Zayandeh-rud Reservoir from the drought mitigation point of view. Furthermore, the proposed model is compared to a stochastic dynamic programming model by investigating different indices such as drought duration, drought severity, drought loss, and reliability of agricultural water demands allocation. The results indicate that the use of the proposed approach can significantly reduce the vulnerability of the system during hydrological droughts and increases the long-term benefits of agricultural water demand allocation.     

Reservoir Inflow Modeling Using Temporal Neural Networks with Forgetting Factor Approach

In this paper, a recursive training procedure with forgetting factor is proposed for on-line calibration of temporal neural networks. The forgetting factor discounts old measurements through an on-line model calibration. The forgetting factor approach enables the recursive algorithm to reduce the effect of the older error data by multiplying the error data by a discounting factor. The proposed procedure is used to calibrate a temporal neural network for reservoir inflow modeling. The mean monthly inflow of the Karoon-III reservoir dam in the south-western part of Iran is used to test the performance of the proposed approach. An autoregressive moving average (ARMA) model is also applied to the same data. The temporal neural network, which is trained with the proposed approach, has shown a significant improvement in the forecast accuracy in comparison with the network trained by the conventional method. It is also demonstrated that the neural network trained with forgetting factor results in better forecasts compared to the statistical ARMA model, which has been calibrated through this approach.     

Long-Term Operation of Irrigation Dams Considering Variable Demands: Case Study of Zayandeh-rud Reservoir, Iran

Dependency of water demands on the climate variation occurs especially in regions where agricultural demand has a significant share of the total water demands. The variability between demands that are based on annual climate conditions may be larger than the uncertainty associated with other explanatory variables in long-term operation of an irrigation dam. This paper illustrates certain benefits of using variable demands for long-term reservoir operation to help manage water resources system in Zayandeh-rud river basin in Iran. A regional optimal allocation of water among different crops and irrigation units is developed. The optimal allocation model is coupled with a reservoir operating model, which is developed based on the certain hedgings that deals with the available water and the water demands mutually. This coupled model is able to activate restrictions on allocating water to agricultural demands considering variation of inflow to the reservoir, variation of demands, and the economic value of allocating water among different crops and irrigation units. Using this model, long-term operation of Zayandeh-rud dam is evaluated considering different scenarios of inflow to the reservoir as well as agricultural demands. The results indicate that the use of operating rules which consider variable demands could significantly improve the efficiency of a water resources system in long-term operation, as it improves the benefit of Zayandeh-rud reservoir operation in comparison with conventional water supply approaches.