عنوان مقاله [English]
Soil erosion and sedimentation are one of the most important environmental , agricultural issues and food production in the world. Soils that are eroded by runoff rain drops are transported to the rivers and released from the catchment areas. Therefore, regional analysis of the sediment load of rivers, especially in arid and semi-arid regions, and its relation to the characteristics of watersheds is very important in estimating the amount of erosion and sediment. Dez Dam Basin is one of the important catchments of the western part of the country in terms of water, energy and agricultural production. Dez river collecting runoff from vast areas of Lorestan, Isfahan, Khuzestan, Chaharmahal Bakhtiari and central provinces, and after watering it into the plains of its route to the Karoon River. The lack of sufficient number of sedimentation stations and the limitation of the number of statistical years as well as the low number of flood sampling during the outbreak of the Dez river indicate the importance of this study in modeling the suspended sediment estimation in the Dez dam basin. Therefore, the aim of this study is to model the relationship between the amount of suspended sediment load with hydrological characteristics and the environment of the dam dam.
Matarials and methods
The catchment area of the Dez River in the middle Zagros Mountains and between the geographic coordinates is 17 ° 31 and 15 ° 34, the northern latitudes are 17 ° 48 and 50 ° 50, respectively. The watershed of the Dez River is considered to be part of the Persian Gulf watershed for the general classification of the watersheds of Iran. The dominant climate in the study area is warm, dry and semi-arid soils. The average annual rainfall of the basin is 670.5 mm. The annual temperature varies from 6.7 to 24.4 degrees Celsius. The Dez River is more than 500 km long in the main drainage of the Dez Basin. For this purpose, 22 sedimentation stations were selected at the upstream of the Dez dam. 16 stations were used to develop suspended sediment estimation models. Then the developed models were validated with the information of 6 other sediment stations. Based on the suggestions of past studies, 15 different physiographic, climatic and hydrological features were considered as input parameters in modeling estimation of suspended sediments. Height, area and basin environment were obtained from maps of 1: 25000 scale with the use of ArcGis9.3 software. The average slope of the watercourses and basins was extracted from the DEM map. The average annual precipitation and average rainfall of rainy and floody days of the year included Azar, Di, Bahman, Esfand, Farvardin and May. Principal Component Analysis (PCA) was used to select the most important parameters. The stations were then divided into two homogeneous groups based on the selected parameters and cluster analysis. Finally, the models were developed step by step using the multivariate regression method. R2, error rate (RE) and mean error (ME) were used to evaluate the models.
Result and discussion
The results showed that the sub-basin of the Takht-Dareh valley with an average of 1256 tons per year is the lowest and the tele-alarm-dose basin has the highest suspended sediment with 8800919 tons per year. Based on the results of the analysis of the main components, four factors of annual discharge with a weight of 94, average annual rainfall with a weight of 88, Horton coefficient with a weight of 90/0 and average height of the basin with a weight of 0.93, the most important parameters Estimates of sediment were suspended. According to the effective factors, sub-basin groups were grouped using cluster analysis. According to the cluster composition at 25, sub-basins were classified into two homogeneous groups. Suspended sediment estimation models were developed based on four factors and using stepwise stepwise multivariable regression method. All developed models in the homogeneous region 1 have more R2 coefficients than those developed in homogeneous region 2. The results of the evaluation of the models showed that the developed model using the four factors with 8% = RE, t ME= 25375 t ya-1, and R2 = 0.93 had the best performance and the least percentage of estimated error. The results also showed that Dubai was the most influential parameter on suspended sediment estimation models annually.
The purpose of this study was to model the relationship between the environmental characteristics of the Dez dam basin and the amount of sediment suspended in this river basin using stepwise multivariate regression method. The results also showed that the regression model derived from all four factors of annual discharge, average height of the basin, horton coefficient and mean rainfall, with the highest determination coefficient (R2 = 0.93), the lowest error (RE = 8%) and the lowest mean error (ME=125375 t ya-1) has a very good performance in estimating suspended sediments. In general, it can be said that the combined use of principal component analysis, cluster analysis and multivariate regression There is an appropriate and acceptable efficiency step in estimating suspended sediments.