نشریه پژوهش های حفاظت آب و خاک
سال شانزدهم شماره 4 (زمستان 1388)

Considering the current trend of ever decreasing water resources and their associated qualitative and quantitative variations, managers are now facing the challenge of offering adaptable planning and allocation decision methods for integrated watershed management. This study focuses on the understanding of the trade-offs among the various outcomes of allocation decisions for Boostan Dam reservoir on the Gorgan-Rud River in Golestan Province of Iran. Control of water level in the reservoir such that enough water is available during dry months to satisfy different needs of the downstream as well as the capability of dam to mitigate floods were considered as the objectives of the study. The recorded flow data suggested that there are four months with high likelihood of flooding events. Sixteen mutually exclusive scenarios were developed to accommodate different and competing water demands of water users in the area. The monthly river flow discharge for one Iranian water-year started from October 2007 was forecasted by a Seasonal Auto Regressive Integrating Moving Average (SARIMA) model. The water allocation analysis was conducted on the basis of the forecasted monthly flow data. To quantify the effects of implementing the scenarios, five indices were selected and quantified. The indices included water storage capacity during flood-prone months, stored water volume at the end of the water-year, the likelihood of overflowing, outflow discharge simulated by a flood routing technique and the risk of dam collapse because of floods. The fuzzy multi-criteria decision making technique was used to trade-off the outcomes and therefore to identify the best allocation scenario. To weight the indices, the process of expert knowledge elicitation was completed using AHP (Analytical Hierarchy Process) technique. The TOPSIS (Technique Ordered Preference by Similarity to the Ideal Solution) model was used to identify the best scenario among the 16 allocation scenarios. The results showed that the scenarios could be categorised in nine different groups among them scenario 7 has been identified as the best allocation scenario followed by scenario1. Scenario 07 was characterized by emphasis on flood control objective for the flood-prone months (May and August) as well as water supply to meet the water demands of the downstream. The integrating approach implemented in this study is capable of assisting the decision makers and dam operators to understand and investigate the possible allocation scenarios, and to recognize the likely outcomes and the trade-offs among them. Thus, the approach supports the decision makers to achieve their objectives capturing different criteria associated with the allocation problems.

The Karun River has very complex problems related to sediments such as formation of big islands, banks failure, huge sediment deposition and reduction of flow velocity and hydraulic capacity. For Karun River training and restoration, many projects and engineering works have been considered to be designed and constructed including levee construction and river dredging. For evaluating of these complex schemes and their interactions, sedimentation studies have to be carried out in this river reach. Mathematical modeling of erosion and deposition of sediments in Karun River from Ahwaz hydraulic station to Farsiat station, a distance of approximately 50 km, has been studied using GSTARS 2.0 model. GSTARS mathematical model using stream tubes concept, has an ability to account for the longitudinal and lateral distribution of sediment deposition in the rivers. For hydraulic and sediment calibration of this mathematical model, historical water surface elevation of flood in 2005 and cross section topography between 1998 and 2005, have been used, respectively. Calibration results showed that Ackers-White (1974) and Toffaleti (1969) sediment transport equations had good agreement with field data.

Landslide is one of the natural hazards that makes numerous financial and life damages each year. Using landslide hazard zonation, one can detect susceptible and high potential landslide hazard areas. In this study efficiency of four landslide hazard zonation models including empirical revised Moravarson model, nonweighted Density Area model, weighted Density Area model, and multivariate regression model have been compared at a part of Alamout watershed in general level (1/50000 scale). Landslide distribution map of area was first prepared using air photo interpretations and field studies. After reviewing available resources along with reviewing the benefit of experts, eight parameters including (lithology, slope, aspect, height, distance to fault, land use, rain and earthquake acceleration) were selected as landslide effective factors. Potential landslide hazard maps were also provided with four models using geographic information system (GIS) and crossing layers in Arc GIS software. AHP and pair comparing technique were used for weighting the parameters in weighted Density Area model. This technique was also used for quantifying the qualitative categories of land use, aspect, and lithology parameters in multivariate regression model. Chi square test was used for comparing the difference between hazard classes of models and sum of quality index (Qs) was used for efficiency of models. Results showed that measured chi square rate is meaningful at 99% of confidence interval, and there is suitable dissociation among landslide hazard classes. Three factors including aspect, rain and earthquake acceleration could not be entered to model showing, low correlation between three parameters and occurred landslides. Multivariate regression model was also selected for Alamoute watershed with Qs equal to 0.947.

Water resource management, especially in groundwater, is very important in arid and semiarid regions. Natural and anthropogenic phenomena have caused critical conditions and fall in groundwater table within last decades in many places of country such as Khorasan Razavi Province. Therefore for assessment of falling of groundwater table, Mashhad Plain, which is one of the important and largest plain in province, was chosen. To do this research, 70 piezometers well were studied during 2 ten-years periods (1987-97 and 1997-2007). Statistical data were first collected and arranged. Afterwards, isopieses and classification of groundwater level were performed by interpolation method in GIS environment using geology, topography and mountain-plain limit maps. Also water level hydrograph were drawn in Excel software. The result of classification of groundwater level maps indicated that groundwater table decreased about 30 meters in central and western regions of aquifer. The concentration of exploitation wells with high debits in this region, has confirmed former findings according to the classification of exploitation map. According to hydrographs, groundwater table has decreased to 12.1 meters during 20 years, 60 cm water level fall annually. Drought, overexploitation, growth of population, increase in agricultural lands and increasing number of wells are the most important factors in decreasing of groundwater table.

Using two different artificial neural methods [Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS)] and two empirical evapotranspiration models [Penman-Montieth FAO56 (PMF56) and Blaney-Cridle (BC)] daily reference evapotranspiration values (ET0) were estimated. Time series of two years observed meteorological data (1997-1998) were used as the input and output for modeling non-linear system of ET0 values. The structures of ANN and ANFIS network were constructed in such a way that in addition to their ability for comparing the performance of different arrays, they were able to show the effect of dynamical behavior of the used networks. The model outputs were validated against a two years (1997-1998) Lysimeter (1*1*2.25m) data, in addition to the daily weather observations (minimum and maximum temperatures, minimum and maximum relative humidity, sunshine duration, wind speed) in cold-semi climate of Hamedan. The results indicate that neural networks perform reliable ET0 estimations in both decreasing and increasing steps, in comparison with the classic methods. In addition, the neural intelligent networks provide more accurate estimates in shorter computer time compared with the results obtained by classic approaches. It is shown that validations against independent data set are reliable (with coefficient of correlation of about 0.95) for such non-linear dynamical processes. The comparisons of ANN and ANFIS results indicate that ANN performs more accurate ET0 estimates. For the climate condition of Hamedan, the ANN method with 6-3-1 arrays and feed-forward back propagation (FFBP) learning rule performed better predictions compared to the ET0 estimated by ANFIS.

The aim of this research is to investigate the existence and the three dimensional distribution of the pollution around Sarkhoun refinery. Following the initial investigation of the site, wells were dug up to hardpans followed by sampling. Soil pollution mappings and its distribution were performed by Surfer and Arc GIS and MATLAB softwares. The results showed that due to the lack of underground water in the area and the hardpans near soil surface, pollution spread has happened through gas spread from the limited region above the hardpan. Geostatistic studies and 3-dimential estimated pollutions mapping by software showed the range of studied polluted soil for TOG and TPH was 3633 and 2612 mg/Kg near burning pits and to 7 and 4 mg/Kg in the west south of the refinery. Furthermore, pollution spread in the direction of soil layers gradient is towards Salt River, which is placed in the east of the refinery. In the east of refinery, sudden increase in both pollution maps was related to the tube lines breakage. The result showed that TOG pollution map and TPH pollution map had similar trend. The results of the simulations revealed that in Sarkhoun soils by increasing in water debit to burning pits, risk of pollution spreading increased, Therefore appropriate remedy for blockage of spreading would be necessary.

Tillage and organic matter management have some impact on soil physical, chemical and biological properties, therefore they influence water and solute transport in soil. This study was performed in two consecutive years. Different tillage treatments including disk harrowing + two disk harrowings as reduced tillage (15 cm plowing depth) and moldboard plowing + two disk harrowings as conventional tillage system (30 cm plowing depth) were used. Also, three levels of 0, 30 and 60 tons per hectare farmyard manure were also used as organic manure treatments. The above mentioned treatments were done in a split block design with 3 replications and under corn cultivation for two years. In order to determine solute transport, potassium bromide was added to soil at concentration of 16.67 g/l. Soil sampling from furrow was conducted after 126, 315 and 630 mm irrigation, respectively. Sampling depth were 0-15, 15-30, 30-60, 60-90 and 90-120 cm. The results showed that bromide leaching and transport was significantly lower under reduced tillage (T1) in comparison to conventional tillage (T2) irrespective to water irrigation content. This is due to higher soil loosing by T2 compared to T1. The results indicated that manure had no significant effect on bromide transport. With increasing irrigation water content, bromide leaching to lower layers increased as well. Under T1, after 126 and 315 mm water, most of bromide has remained at depth of 0-15 cm. However, by adding 630 mm of water, bromide concentration reduced considerably at surface layer. Although bromide leaching was higher at T2 than T1 treatment, but due to low permeability of subsoil layers, bromide concentration (tracer) at topsoil was considerable. But, by adding 630 mm water in the last stage of sampling, the bromide concentration at surface layer (0-15 cm) showed a reduction, therefore a great amount of trace element has escaped from this layer.

Climate change is one of the modern problems for human beings. It is a disaster for the earth. The recent climate changes are mostly influenced by increase in amount of CO2 and some greenhouse gases in atmosphere. To consider the climate change, data of meteorological station (synoptic) in Urmia from 1950-2000 has been used. The amount of changes in parameters related to temperature, precipitation and number of sunny hours has been compared through linear regression. The results showed that maximum average temperature within last fifty years has not been uniformly increased (P=0.05). However, minimum average temperature decreased until 1978-83 in each decade during the following months like: March and October (0/6ºC), June, April and November (0/4ºC), May and August (0.5ºC), July (0.3ºC), September (0.7ºC). There has been (0.5ºC) increase annually. The minimum average temperature increased within 1978-83 as the following: June (1.4ºC), July (1.1ºC), August (1.5ºC), September (0.9ºC) and December (1.5ºC). Therefore, there has been (1.1ºC) annual increase. Maximum temperature decreased in each decade as the following: July (0.4ºC), August (0.5ºC) and October (0.6ºC). There has been a decrease of (0.4ºC) annually. Average amount of rainfall was not significant during last 50 years. Average numbers of sunny hours in each decade increased as February (1.475 hour), March (1.83 hour) and May (1.54 hour). Consequently, It is said that universal climate change in this region resulted in temperature decrease and did not have any effect on rainfall and numbers of sunny hours.

In the operation of water resources system, adequate continuous rainfall data are required as inputs into the reservoirs. If historical records are limited, they are extended by generating synthetic rainfall sequences with longer records. Stochastic monthly and annual rainfall sequences simulation is considered in this paper which tries to generate data with long-term dependence and preserve historical records distribution function. Annual and monthly rainfalls for Dargaz station with 15 years records as representative rainfall station for Illanjogh dam in the north of Khorasan province were satisfactorily simulated by stochastic models. A first-order Markov process is used for annual data, whereas the method of fragment is used for monthly data. Comparison of the simulated, and historical data, and statistical similarity of the annual and monthly models compared with the historical one, confirmed the correct performance of the annual and monthly rainfall production model. Considering present conditions in this problem (short- term statistical period of rainfall with high coefficient of variation) showed that fragment method took an acceptable result such that could confirm the advantage of its usage in undesirable conditions.