Author(s): Jie Qin; Jochen Aberle; Pierre-Yves Henry; Teng Wu; Deyu Zhong
Keywords: Armoured gravel surfaces; bed roughness; Monte Carlo method; morphodynamics; roughness scales
Abstract: Structure function analyses are increasingly used to investigate spatial correlation patterns of river bed roughness. However, traditional structure function analyses remain mostly qualitative as the statistical significance of correlation patterns has not been explicitly addressed. This paper proposes a methodology for the analysis of 2D second-order structure functions based on statistical significance testing. Using data from flume experiments with gravel beds, a method to create surrogate rough beds is proposed. The surrogate beds are subsequently used to carry out Monte-Carlo simulations which in turn are used to evaluate statistical significance of the structure functions values. The method is applied to two datasets of gravel surfaces and the results are used to investigate the physical meaning of the obtained significance patterns. The results confirm previous studies that the orientation of large grains can be described by 2D second-order structure functions through a central ellipse reflecting areas of high spatial correlation. The main axes of the ellipses can hence be interpreted to define characteristic roughness length scales. The novelty of the present paper is that the proposed significance testing allows for a more rigorous definition of the spatial extent of the central ellipse.