عنوان مقاله [English]
نویسندگان [English]چکیده [English]
Estimating the correct volume of sediment in alluvial rivers plays a crucial role in river engineering, water resources, structures, facilities, water and environmental projects. Using the observed records of sediment load is the most reliable way in estimating the volume of sediment. Because of user unfriendly measurement tools and its remarkable fluctuation within the river section, sediment sampling is really difficult and needs highly experienced operators. In addition, it is a costly and time consuming practice. Because of these limitations, frequency of sediment observation has decreased around the world, especially in developing countries and remote areas. Therefore, researchers considered some modeling approaches based on diverse terrain attributes and hydrological variables. Some physically-based models such as the unit stream power (USP) theory of Yang, the SHESED model of Wicks and Bathurst are able to universally predict sediment yield of a watershed, but these models need lots of detailed information such as geological, hydraulic, and hydrological characteristics of the river basin that make them complex and costly. There is similar condition for process-based models such as the modified universal soil loss equation (MUSLE), introduced by Williams, and its family (USLE and RUSLE). Therefore, some simpler models were developed by researchers which are based on many simplifying assumptions and empirical relationships, especially for rainfall and runoff erosive effects. These models can be employed in areas with few data that result in high uncertain results. Application of data-driven models such as different techniques of artificial intelligence are the alternative method to the previous models. Artificial neural network (ANN) is one of the most famous and strong data-driven technique that has been proved to be practical in modeling non-linear systems or complicated hydrologic processes such as sediment transport. Also, conventional sediment rating curve (SRC) has been widely used all over the world. Although these methods might be similar in some aspects, but each of them has its own characteristics that make them different from the others. In the present study, we use combinations of the above tools in a new approach and a hybrid model was developed to utilize the quality of each method in estimating the bed load sediment in the Yazdakan gauging station, simultaneously.
For the present study, Yazdakan Bridge gauging site (38.476724o, 44.798362o) located on Ghotorchay River was studied. The Ghotorchay River located in the Aras catchment with an area of 3471.9 km2 is a river in West Azerbaijan province of Iran. It arises in the northwest of the Zagros Mountain range in Turkey. The total length of Ghotorchay main streams in Turkey before entering to Iran reaches to 115 kilometers. It flows from the border of Iran and Turkey at the height of 290 m from west to east through lands with steep slopes and passes the length of about 70 kilometers to reach to the Khoy city of West Azarbaijan. After leaving Khoy, the river turns north and flows through Evaghly valley where joins to the Zilber Chay river and extends until meets the Agh Chay river at Marakan county. The river with its first name on Farhadi border outpost connects to the Aras river.
For current study, 133 paired data of flow discharge and corresponding bed load sediment were considered from the Yazdakan Bridge gauging site. The duration of collecting data started from January of 1999 to September of 2009 that measured by West Azarbaijan regional water authority. Monthly data of bed load sediment and discharge related to the time of the experiment were recorded. For constructing each model the data set was divided into two groups which the first 25% of paired date as group one was utilized for testing and the second group, including 75% of paired data employed for both training and validation of models.
In this study, discharge flow data were utilized as input for each of ANN and the sediment rating curve (SRC) techniques, separately to estimate bed load sediment. An idea to optimize the results of the SRC by ANN, navigate the author to a hybrid model made from both of the techniques which called HAS in a way that the error of SRC was predicted using ANN.
Finally obtained results demonstrated that the results of HAS model is in good agreement with the observed bed load sediment concentration values; which depict better results than SRC and ANN models. For example, the calculated root of mean square error for HAS model is 213.44 ton/day, while it is 262.028 tons/day and 238.305 tons/day for ANN and SRC models, respectively. In general, it is illustrated from the results that the HAS hybrid model presents better performance than SRC and ANN for estimating bed load sediment in Ghotorchay River.