Prediction of scour pattern around hydraulic structures using geostatistical methods

Abstract

Simulation of sediment movement, identification of scour pattern, and sediment variations play an important role in river rehabilitation and protection of hydraulic structures considering their stability and safety. In general, different shapes of abutments (circular, vertical, and wing wall) and also spur dikes were employed to protect against scouring. The main purpose of this paper is to investigate spatial variations of local scour depth around these hydraulic structures using deterministic spatial estimator known as InverseDistance Weighed (IDW), and two geostatistical approaches known as Ordinary Kriging (OK) and Bayesian Maximum Entropy (BME). To achieve this goal, experimental tests were conducted to measure scour pattern under three types of abutments and spur dikes. To evaluate the results of scour depth, leave-one-out cross-validation technique has been used. The comparison of the computed versus observed scour depth depicts that geostatistical methods (OK and BME) can estimate the maximum scour depth more reliably and accurately than IDW method. Experimental results show that the trend of scour depth for all abutment types does not significantly differ. Maximum scour depths at the wing wall abutment were lower than vertical abutment.