مجله تحقیقات منابع آب ایران
سال دوم شماره 3 (پیاپی 6، زمستان 1385)

In the process of economic development of countries facing population growth and agricultural development, the availability of safe water is of great importance. It is also crucial to consider the social and economical concerns about water exploitation and conveyance. This paper provides an overview of the theoretical issues and operational models for estimating the value and cost of water. The value of water is divided into three major factors: value in use which could be direct or indirect, social benefits, and intrinsic value. The full cost of water is composed of three items: exploitation costs, economic externalities, and environmental externalities. Alavian Dam and its irrigation system is located in the Lake Orumieh Basin. The mean annual release from the dam is 123 million cubic meters, of which around 101 million cubic meters is diverted to the Soufichai irrigation network. According to the results, the full cost of water for the dam and the irrigation network would be 92 and 182 Rial/m3, respectively with an average of 174 Rial/m3. Besides, the sum of both factors of value in use and social value is estimated to be 1622 Rial/m3. Thus the intrinsic value is estimated to be 313 Rial/m3.

As a result of system complexity and incomplete information, the planning process, the “with” and “without” conditions, the project's performance and impacts, and the benefits and costs are never certain. Accordingly, the valuation of uncertainty is a “must”, even on the risk indices as a sub set of uncertainty. This paper focuses on the effects of two main sources of uncertainty; i.e. data and parameter uncertainty, on the risk indices in reservoir simulation and optimization. In reservoir simulation, the effects of uncertainties in simple/disaggregated time series models are studied on the well-known reservoir efficiency indices; i.e. volumetric and occurrence reliabilities. Two additional risk indices; i.e. vulnerability and resilience are also studied. A stochastic dynamic programming model (DPR) is used to study the effects of data continuity uncertainty on reservoir operation rule. In this case the effects of data continuity uncertainty on two indices of Reliability and Losses have been evaluated. Two reservoirs under construction in Fars province, IRAN, have been selected for case study. The results show that the incorporation of uncertainty (i.e. data/parameter uncertainty) in simulation and optimization impact the risk indices differently as the flow variability condition and reservoir storage and operation changes. These results can help the designer to select the appropriate risk indices for reservoir operation.

Drought monitoring is quite important in planning for drought mitigation schemes. Drought indices are normally used for monitoring purposes. However, the indices are calculated at sites and it is required to estimate the spatial distribution of drought in the forms of maps. Geostatistical methods are among such tools that may be applied for preparing spatial distribution maps. In this research, several geostatistical methods including Kriging, Cokriging, TPSS with and without secondary variables as well as weighted moving average (WMA) were assessed with respect to derivation of drought index maps. Drought monitoring and mapping was conducted based on SPI and EDI indices using data of 43 rain gauge stations within Tehran province. The results indicated that although Kriging was the most accurate method, WMA provided reasonable accuracy blended with more efficient computer time, a major consideration in monitoring systems.

The Time–Area method is a suitable technique for watershed routing and can be potentially used as a distributed model. As an advantage, it can also be used as a GIS-based method. The performance of the existing methods for deriving isochrone locations is compared in this study with that of the kinematic wave theorem. In most methods, travel time is proportional to the distance-to-outlet of any point raised to a power. Investigating the numerical value of powers, it is shown that exponent 1.5 of Laurenson’s method yields the closest time-area hydrograph to that of the kinematic wave solution. Therefore, this paper showed that certain empirical isochrone delineation methods could be applicable provided that the travel length is measured towards the outlet.

Limited number of sediment stations and recorded data, together with the inaccurate and insufficient sediment samples in flood events have doubted the river sediment load analysis especially in dry and semi-dry areas. Accordingly the regional analysis seems to be a useful tool aiming appropriate models for estimating sediment loads. In this study the model results are compared for sediment load estimatin in Esfahan and Sirjan watersheds, located in central Iran and comprise 99300 Km2 in area. For 14 sub-watersheds having 11 years of compatible data records the mean daily suspended sediments and 49 physiographic, hydrologic, land use, geologic and climatic characteristics were used respectively as criterion and predictor variables. The multivariable regression techniques were applied with the help of factor analysis, clustering, and discriminant analysis to access the most suitable regression models. The results revealed that two regression models developed for two homogeneous regions could acceptably model the variations. The respective maximum and mean relative error of these models were 35.19 and less than 10%, respectively, obtained from factor analysis technique and selecting one representative variable in every component. The most suitable variables for estimation of mean daily suspended sediment yield of sub-watersheds are the peak discharge with 20 years return period, rangeland percentile, and ruggedness number.

Efforts have been made to estimate the sediment transport and beach profile evolution during storm conditions based on surf zone hydrodynamics. This study compensates, to some extent, for the lack of information describing the coastal processes, particularly cross shore sediment transport and short term beach profile evolution in response to storm events. In this regard, following extensive series of experiments, the Litpack model is used to predict the beach profile evolution in an artificial beach. The profile evolution results predicted by the model compared to those found in the experimental measurements are illustrated clearly. The accuracy of the model is still to be improved; nevertheless, the results illustrate the feasibility of the Litpack model to predict the beach profile evolution in coastal regions. The main objectives of these experiments were to undertake a quantitative comparison of cross shore sediment transport and beach profile evolution with different sizes of sand. Increasing the size of beach material results in reduction in cross shore sediment transport and beach profile evolution for coarse beaches. The results obtained from the model compared to the experimental results are promising and encouraging for further modeling developments. Comparison of the results obtained from experiments and numerical models show that the Litpack program can predict the full scale behavior of beaches with reasonable accuracy, however, due to the scale effects its prediction for the laboratory conditions is not very decisive.

Numerical models are being used widely in river engineering projects. However, the reliability of these models, minimum field data requirement, and the optimal computational processes are of major concern. This study compares the simulated steady flow parameters from a well-known numerical models with that of a true river physical model. A fixed-bed model is developed for the Nazloo River, IRAN, with horizontal scale of 1:100 and vertical scale of 1:20. Results from HEC-RAS (1D), BRI-STARS (quasi 2D), and FAST-2D (2D) were compared for corresponding flow parameters under similar initial and boundary conditions. This was made for five different flows in three sub-reaches along the river (i.e. downstream of a bridge, bridge reach, upstream of the bridge). The best adapted results were produced by HEC-RAS. BRI-STARS and FAST-2D models took the next ranks respectively. The averaged predictive errors were rated 1.8, 2.2, and 5.5 percent for water surface elevations and 10.2, 9, and 12.6 percent for mean velocities, respectively. The application of each model is recommended in river projects subject to the inclusion of the degree of certainties provided by the present study.

In hydrologic regional frequency analysis, information of one or more stations are generalized for a region. In this case, the variation of this information is assumed to have a defined coefficient called scale. In this study, scaling properties of 7-day low flow of 19 basins in Mazandaran province in northern Iran were investigated. Scaling implies the relationship between probabilistic features and the area of the basin. If this exponent (called scaling exponent) has a linear relationship with drainage area and does not include a significant relationship with probability weighted moments, low flow is called “simple scaling”. The study showed that the logarithm of the drainage area and the logarithm of the probability weighted moments of order 1 to 10 are linear and the scaling exponent did not have a significant relationship with probability weighted moments. Acordingly the 7-day low flows of Mazandaran province are simple scaling.