Spectral quantification of nonlinear behaviour of the nearshore seabed and correlations with potential forcings at Duck, N.C., U.S.A.
Rebecca B Hoyle
Dominic E Reeve
- Publisher: Public Library of Science (PLoS)
(issn: 1932-6203, eissn: 1932-6203)
Social and Behavioral Sciences | Nonlinear Sciences - Chaotic Dynamics | Geomorphology | Research Article | Earth Sciences | Mathematics | Coastal Engineering | Geophysics | Medicine | Engineering | Q | Marine and Aquatic Sciences | R | Physical Oceanography | Civil Engineering | Geography | Oceanography | Physics | Science | Erosion | Nonlinear Dynamics
Local bathymetric quasi-periodic patterns of oscillation are identified from monthly profile surveys taken at two shore-perpendicular transects at the USACE field research facility in Duck, North Carolina, USA, spanning 24.5 years and covering the swash and surf zones. The chosen transects are the two furthest (north and south) from the pier located at the study site. Research at Duck has traditionally focused on one or more of these transects as the effects of the pier are least at these locations. The patterns are identified using singular spectrum analysis (SSA). Possible correlations with potential forcing mechanisms are discussed by 1) doing an SSA with same parameter settings to independently identify the quasi-periodic cycles embedded within three potentially linked sequences: monthly wave heights (MWH), monthly mean water levels (MWL) and the large scale atmospheric index known as the North Atlantic Oscillation (NAO) and 2) comparing the patterns within MWH, MWL and NAO to the local bathymetric patterns. The results agree well with previous patterns identified using wavelets and confirm the highly nonstationary behaviour of beach levels at Duck; the discussion of potential correlations with hydrodynamic and atmospheric phenomena is a new contribution. The study is then extended to all measured bathymetric profiles, covering an area of 1100m (alongshore) by 440m (cross-shore), to 1) analyse linear correlations between the bathymetry and the potential forcings using multivariate empirical orthogonal functions (MEOF) and linear correlation analysis and 2) identify which collective quasi-periodic bathymetric patterns are correlated with those within MWH, MWL or NAO, based on a (nonlinear) multichannel singular spectrum analysis (MSSA). (...continued in submitted paper)