Distinguishing one-way from coupled dynamical interactions in nearshore morphodynamics
Kenneth D Ells1, Dylan McNamara2, Nicholas Cortale1 and Derek Jeffrey Grimes3, (1)UNC Wilmington, Wilmington, NC, United States, (2)University of North Carolina Wilmington, Wilmington, NC, United States, (3)University of California, San Diego, Scripps Institution of Oceanography, La Jolla, CA, United States
Abstract:
Understanding nearshore morphodynamics is critical for management of coastal economies and ecosystems, as well as for reconstruction of paleoenvironmental signals. Yet a hallmark of environmental systems is their inherent complexity. Common tools for nearshore research range from simple linear extrapolation to heavily parameterized numerical models based on conservation principles. Model-dependent statistical approaches often render the distinction between causation and correlation unclear, while the parameterizations of dynamical models tend to cloud insight and complicate comparisons with nature. These drawbacks indicate the need for techniques that can determine the degrees to which system variables are linked by one-way forcing or two-way coupling. Here we apply recently developed time series analysis methods for detecting dynamical linkages in complex systems to nearshore morphodynamics using a recently deployed coastal video monitoring system located at Wrightsville Beach, NC, USA.
Recent work using nonlinear forecasting techniques suggests that local nonlinear interactions tend to affect daily intertidal profile adjustments, where hourly time lapsed images were used in conjunction with wave and tide data to reconstruct daily foreshore profiles (Grimes et al. in press). To determine the degree to which autogenic dynamics in the foreshore dominate over external forcing, we explore a method proposed by Sugihara et al., (2012) to distinguish “driving” in coupled dynamic systems by extending nonlinear state-space reconstruction. Results will be further compared with other information-theoretic methods for measuring information transfer (e.g. Schreiber, 2000). Our monitoring system has captured the most recent in a long history of beach nourishment projects at Wrightsville Beach, allowing us to extend our analysis to dynamical regimes both before and after a large-scale environmental disturbance. It is also capable of distinguishing a range of beach characteristics, including beach cusps, rip currents, and human occupancy, and results will be presented that explore dynamical connections between these various features.
